♻ ☆ RoboDuet: A Framework Affording Mobile-Manipulation and Cross-Embodiment

Guoping Pan, Qingwei Ben, Zhecheng Yuan, Guangqi Jiang, Yandong Ji, Jiangmiao Pang, Houde Liu, Huazhe Xu

Combining the mobility of legged robots with the manipulation skills of arms has the potential to significantly expand the operational range and enhance the capabilities of robotic systems in performing various mobile manipulation tasks. Existing approaches are confined to imprecise six degrees of freedom (DoF) manipulation and possess a limited arm workspace. In this paper, we propose a novel framework, RoboDuet, which employs two collaborative policies to realize locomotion and manipulation simultaneously, achieving whole-body control through interactions between each other. Surprisingly, going beyond the large-range pose tracking, we find that the two-policy framework may enable cross-embodiment deployment such as using different quadrupedal robots or other arms. Our experiments demonstrate that the policies trained through RoboDuet can accomplish stable gaits, agile 6D end-effector pose tracking, and zero-shot exchange of legged robots, and can be deployed in the real world to perform various mobile manipulation tasks. Our project page with demo videos is at https://locomanip-duet.github.io .

♻ ☆ Leveraging Symmetry in RL-based Legged Locomotion Control

Zhi Su, Xiaoyu Huang, Daniel Ordoñez-Apraez, Yunfei Li, Zhongyu Li, Qiayuan Liao, Giulio Turrisi, Massimiliano Pontil, Claudio Semini, Yi Wu, Koushil Sreenath

Model-free reinforcement learning is a promising approach for autonomously solving challenging robotics control problems, but faces exploration difficulty without information of the robot's kinematics and dynamics morphology. The under-exploration of multiple modalities with symmetric states leads to behaviors that are often unnatural and sub-optimal. This issue becomes particularly pronounced in the context of robotic systems with morphological symmetries, such as legged robots for which the resulting asymmetric and aperiodic behaviors compromise performance, robustness, and transferability to real hardware. To mitigate this challenge, we can leverage symmetry to guide and improve the exploration in policy learning via equivariance/invariance constraints. In this paper, we investigate the efficacy of two approaches to incorporate symmetry: modifying the network architectures to be strictly equivariant/invariant, and leveraging data augmentation to approximate equivariant/invariant actor-critics. We implement the methods on challenging loco-manipulation and bipedal locomotion tasks and compare with an unconstrained baseline. We find that the strictly equivariant policy consistently outperforms other methods in sample efficiency and task performance in simulation. In addition, symmetry-incorporated approaches exhibit better gait quality, higher robustness and can be deployed zero-shot in real-world experiments.

♻ ★ Learning Quadruped Locomotion Using Differentiable Simulation

Yunlong Song, Sangbae Kim, Davide Scaramuzza

While most recent advancements in legged robot control have been driven by model-free reinforcement learning, we explore the potential of differentiable simulation. Differentiable simulation promises faster convergence and more stable training by computing low-variant first-order gradients using the robot model, but so far, its use for legged robot control has remained limited to simulation. The main challenge with differentiable simulation lies in the complex optimization landscape of robotic tasks due to discontinuities in contact-rich environments, e.g., quadruped locomotion. This work proposes a new, differentiable simulation framework to overcome these challenges. The key idea involves decoupling the complex whole-body simulation, which may exhibit discontinuities due to contact, into two separate continuous domains. Subsequently, we align the robot state resulting from the simplified model with a more precise, non-differentiable simulator to maintain sufficient simulation accuracy. Our framework enables learning quadruped walking in minutes using a single simulated robot without any parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills, including trot, pace, bound, and gallop, on challenging terrains in minutes. Additionally, our policy achieves robust locomotion performance in the real world zero-shot. To the best of our knowledge, this work represents the first demonstration of using differentiable simulation for controlling a real quadruped robot. This work provides several important insights into using differentiable simulations for legged locomotion in the real world.

♻ ☆ Scalable Non-Cartesian Magnetic Resonance Imaging with R2D2

Yiwei Chen, Chao Tang, Amir Aghabiglou, Chung San Chu, Yves Wiaux

We propose a new approach for non-Cartesian magnetic resonance image reconstruction. While unrolled architectures provide robustness via data-consistency layers, embedding measurement operators in Deep Neural Network (DNN) can become impractical at large scale. Alternative Plug-and-Play (PnP) approaches, where the denoising DNNs are blind to the measurement setting, are not affected by this limitation and have also proven effective, but their highly iterative nature also affects scalability. To address this scalability challenge, we leverage the "Residual-to-Residual DNN series for high-Dynamic range imaging (R2D2)" approach recently introduced in astronomical imaging. R2D2's reconstruction is formed as a series of residual images, iteratively estimated as outputs of DNNs taking the previous iteration's image estimate and associated data residual as inputs. The method can be interpreted as a learned version of the Matching Pursuit algorithm. We demonstrate R2D2 in simulation, considering radial k-space sampling acquisition sequences. Our preliminary results suggest that R2D2 achieves: (i) suboptimal performance compared to its unrolled incarnation R2D2-Net, which is however non-scalable due to the necessary embedding of NUFFT-based data-consistency layers; (ii) superior reconstruction quality to a scalable version of R2D2-Net embedding an FFT-based approximation for data consistency; (iii) superior reconstruction quality to PnP, while only requiring few iterations.

comment: submitted to IEEE EUSIPCO 2024

♻ ☆ Generative 3D Part Assembly via Part-Whole-Hierarchy Message Passing

Bi'an Du, Xiang Gao, Wei Hu, Renjie Liao

Generative 3D part assembly involves understanding part relationships and predicting their 6-DoF poses for assembling a realistic 3D shape. Prior work often focus on the geometry of individual parts, neglecting part-whole hierarchies of objects. Leveraging two key observations: 1) super-part poses provide strong hints about part poses, and 2) predicting super-part poses is easier due to fewer superparts, we propose a part-whole-hierarchy message passing network for efficient 3D part assembly. We first introduce super-parts by grouping geometrically similar parts without any semantic labels. Then we employ a part-whole hierarchical encoder, wherein a super-part encoder predicts latent super-part poses based on input parts. Subsequently, we transform the point cloud using the latent poses, feeding it to the part encoder for aggregating super-part information and reasoning about part relationships to predict all part poses. In training, only ground-truth part poses are required. During inference, the predicted latent poses of super-parts enhance interpretability. Experimental results on the PartNet dataset show that our method achieves state-of-the-art performance in part and connectivity accuracy and enables an interpretable hierarchical part assembly. Code is available at https://github.com/pkudba/3DHPA.

♻ ☆ Language Models are Free Boosters for Biomedical Imaging Tasks

Zhixin Lai, Jing Wu, Suiyao Chen, Yucheng Zhou, Naira Hovakimyan

In this study, we uncover the unexpected efficacy of residual-based large language models (LLMs) as part of encoders for biomedical imaging tasks, a domain traditionally devoid of language or textual data. The approach diverges from established methodologies by utilizing a frozen transformer block, extracted from pre-trained LLMs, as an innovative encoder layer for the direct processing of visual tokens. This strategy represents a significant departure from the standard multi-modal vision-language frameworks, which typically hinge on language-driven prompts and inputs. We found that these LLMs could boost performance across a spectrum of biomedical imaging applications, including both 2D and 3D visual classification tasks, serving as plug-and-play boosters. More interestingly, as a byproduct, we found that the proposed framework achieved superior performance, setting new state-of-the-art results on extensive, standardized datasets in MedMNIST-2D and 3D. Through this work, we aim to open new avenues for employing LLMs in biomedical imaging and enriching the understanding of their potential in this specialized domain.

♻ ☆ Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving CVPR 2024

Junhao Zheng, Chenhao Lin, Jiahao Sun, Zhengyu Zhao, Qian Li, Chao Shen

Deep learning-based monocular depth estimation (MDE), extensively applied in autonomous driving, is known to be vulnerable to adversarial attacks. Previous physical attacks against MDE models rely on 2D adversarial patches, so they only affect a small, localized region in the MDE map but fail under various viewpoints. To address these limitations, we propose 3D Depth Fool (3D$^2$Fool), the first 3D texture-based adversarial attack against MDE models. 3D$^2$Fool is specifically optimized to generate 3D adversarial textures agnostic to model types of vehicles and to have improved robustness in bad weather conditions, such as rain and fog. Experimental results validate the superior performance of our 3D$^2$Fool across various scenarios, including vehicles, MDE models, weather conditions, and viewpoints. Real-world experiments with printed 3D textures on physical vehicle models further demonstrate that our 3D$^2$Fool can cause an MDE error of over 10 meters.

comment: Accepted by CVPR 2024

♻ ☆ Coarse-Tuning for Ad-hoc Document Retrieval Using Pre-trained Language Models LREC

Atsushi Keyaki, Ribeka Keyaki

Fine-tuning in information retrieval systems using pre-trained language models (PLM-based IR) requires learning query representations and query-document relations, in addition to downstream task-specific learning. This study introduces coarse-tuning as an intermediate learning stage that bridges pre-training and fine-tuning. By learning query representations and query-document relations in coarse-tuning, we aim to reduce the load of fine-tuning and improve the learning effect of downstream IR tasks. We propose Query-Document Pair Prediction (QDPP) for coarse-tuning, which predicts the appropriateness of query-document pairs. Evaluation experiments show that the proposed method significantly improves MRR and/or nDCG@5 in four ad-hoc document retrieval datasets. Furthermore, the results of the query prediction task suggested that coarse-tuning facilitated learning of query representation and query-document relations.

comment: Accepted at LREC-COLING 2024

♻ ☆ Imitating Cost-Constrained Behaviors in Reinforcement Learning

Qian Shao, Pradeep Varakantham, Shih-Fen Cheng

Complex planning and scheduling problems have long been solved using various optimization or heuristic approaches. In recent years, imitation learning that aims to learn from expert demonstrations has been proposed as a viable alternative to solving these problems. Generally speaking, imitation learning is designed to learn either the reward (or preference) model or directly the behavioral policy by observing the behavior of an expert. Existing work in imitation learning and inverse reinforcement learning has focused on imitation primarily in unconstrained settings (e.g., no limit on fuel consumed by the vehicle). However, in many real-world domains, the behavior of an expert is governed not only by reward (or preference) but also by constraints. For instance, decisions on self-driving delivery vehicles are dependent not only on the route preferences/rewards (depending on past demand data) but also on the fuel in the vehicle and the time available. In such problems, imitation learning is challenging as decisions are not only dictated by the reward model but are also dependent on a cost-constrained model. In this paper, we provide multiple methods that match expert distributions in the presence of trajectory cost constraints through (a) Lagrangian-based method; (b) Meta-gradients to find a good trade-off between expected return and minimizing constraint violation; and (c) Cost-violation-based alternating gradient. We empirically show that leading imitation learning approaches imitate cost-constrained behaviors poorly and our meta-gradient-based approach achieves the best performance.

comment: Accepted to the 34th International Conference on Automated Planning and Scheduling (ICAPS-24)

♻ ☆ Re2LLM: Reflective Reinforcement Large Language Model for Session-based Recommendation

Ziyan Wang, Yingpeng Du, Zhu Sun, Haoyan Chua, Kaidong Feng, Wenya Wang, Jie Zhang

Large Language Models (LLMs) are emerging as promising approaches to enhance session-based recommendation (SBR), where both prompt-based and fine-tuning-based methods have been widely investigated to align LLMs with SBR. However, the former methods struggle with optimal prompts to elicit the correct reasoning of LLMs due to the lack of task-specific feedback, leading to unsatisfactory recommendations. Although the latter methods attempt to fine-tune LLMs with domain-specific knowledge, they face limitations such as high computational costs and reliance on open-source backbones. To address such issues, we propose a Reflective Reinforcement Large Language Model (Re2LLM) for SBR, guiding LLMs to focus on specialized knowledge essential for more accurate recommendations effectively and efficiently. In particular, we first design the Reflective Exploration Module to effectively extract knowledge that is readily understandable and digestible by LLMs. To be specific, we direct LLMs to examine recommendation errors through self-reflection and construct a knowledge base (KB) comprising hints capable of rectifying these errors. To efficiently elicit the correct reasoning of LLMs, we further devise the Reinforcement Utilization Module to train a lightweight retrieval agent. It learns to select hints from the constructed KB based on the task-specific feedback, where the hints can serve as guidance to help correct LLMs reasoning for better recommendations. Extensive experiments on multiple real-world datasets demonstrate that our method consistently outperforms state-of-the-art methods.

comment: 11 pages, 4 figures

♻ ☆ MA4DIV: Multi-Agent Reinforcement Learning for Search Result Diversification

Yiqun Chen, Jiaxin Mao, Yi Zhang, Dehong Ma, Long Xia, Jun Fan, Daiting Shi, Zhicong Cheng, Simiu Gu, Dawei Yin

The objective of search result diversification (SRD) is to ensure that selected documents cover as many different subtopics as possible. Existing methods primarily utilize a paradigm of "greedy selection", i.e., selecting one document with the highest diversity score at a time. These approaches tend to be inefficient and are easily trapped in a suboptimal state. In addition, some other methods aim to approximately optimize the diversity metric, such as $\alpha$-NDCG, but the results still remain suboptimal. To address these challenges, we introduce Multi-Agent reinforcement learning (MARL) for search result DIVersity, which called MA4DIV. In this approach, each document is an agent and the search result diversification is modeled as a cooperative task among multiple agents. This approach allows for directly optimizing the diversity metrics, such as $\alpha$-NDCG, while achieving high training efficiency. We conducted preliminary experiments on public TREC datasets to demonstrate the effectiveness and potential of MA4DIV. Considering the limited number of queries in public TREC datasets, we construct a large-scale dataset from industry sources and show that MA4DIV achieves substantial improvements in both effectiveness and efficiency than existing baselines on a industrial scale dataset.

♻ ☆ World Models via Policy-Guided Trajectory Diffusion

Marc Rigter, Jun Yamada, Ingmar Posner

World models are a powerful tool for developing intelligent agents. By predicting the outcome of a sequence of actions, world models enable policies to be optimised via on-policy reinforcement learning (RL) using synthetic data, i.e. in "in imagination". Existing world models are autoregressive in that they interleave predicting the next state with sampling the next action from the policy. Prediction error inevitably compounds as the trajectory length grows. In this work, we propose a novel world modelling approach that is not autoregressive and generates entire on-policy trajectories in a single pass through a diffusion model. Our approach, Policy-Guided Trajectory Diffusion (PolyGRAD), leverages a denoising model in addition to the gradient of the action distribution of the policy to diffuse a trajectory of initially random states and actions into an on-policy synthetic trajectory. We analyse the connections between PolyGRAD, score-based generative models, and classifier-guided diffusion models. Our results demonstrate that PolyGRAD outperforms state-of-the-art baselines in terms of trajectory prediction error for short trajectories, with the exception of autoregressive diffusion. For short trajectories, PolyGRAD obtains similar errors to autoregressive diffusion, but with lower computational requirements. For long trajectories, PolyGRAD obtains comparable performance to baselines. Our experiments demonstrate that PolyGRAD enables performant policies to be trained via on-policy RL in imagination for MuJoCo continuous control domains. Thus, PolyGRAD introduces a new paradigm for accurate on-policy world modelling without autoregressive sampling.

comment: Published in TMLR, March 2024

♻ ☆ LLMs Are Few-Shot In-Context Low-Resource Language Learners

Samuel Cahyawijaya, Holy Lovenia, Pascale Fung

In-context learning (ICL) empowers large language models (LLMs) to perform diverse tasks in underrepresented languages using only short in-context information, offering a crucial avenue for narrowing the gap between high-resource and low-resource languages. Nonetheless, there is only a handful of works explored ICL for low-resource languages with most of them focusing on relatively high-resource languages, such as French and Spanish. In this work, we extensively study ICL and its cross-lingual variation (X-ICL) on 25 low-resource and 7 relatively higher-resource languages. Our study not only assesses the effectiveness of ICL with LLMs in low-resource languages but also identifies the shortcomings of in-context label alignment, and introduces a more effective alternative: query alignment. Moreover, we provide valuable insights into various facets of ICL for low-resource languages. Our study concludes the significance of few-shot in-context information on enhancing the low-resource understanding quality of LLMs through semantically relevant information by closing the language gap in the target language and aligning the semantics between the targeted low-resource and the high-resource language that the model is proficient in. Our work highlights the importance of advancing ICL research, particularly for low-resource languages.

♻ ☆ Dial-MAE: ConTextual Masked Auto-Encoder for Retrieval-based Dialogue Systems NAACL 2024

Zhenpeng Su, Xing Wu, Wei Zhou, Guangyuan Ma, Songlin Hu

Dialogue response selection aims to select an appropriate response from several candidates based on a given user and system utterance history. Most existing works primarily focus on post-training and fine-tuning tailored for cross-encoders. However, there are no post-training methods tailored for dense encoders in dialogue response selection. We argue that when the current language model, based on dense dialogue systems (such as BERT), is employed as a dense encoder, it separately encodes dialogue context and response, leading to a struggle to achieve the alignment of both representations. Thus, we propose Dial-MAE (Dialogue Contextual Masking Auto-Encoder), a straightforward yet effective post-training technique tailored for dense encoders in dialogue response selection. Dial-MAE uses an asymmetric encoder-decoder architecture to compress the dialogue semantics into dense vectors, which achieves better alignment between the features of the dialogue context and response. Our experiments have demonstrated that Dial-MAE is highly effective, achieving state-of-the-art performance on two commonly evaluated benchmarks.

comment: This paper has been accepted by NAACL 2024

♻ ☆ Scalable Non-Cartesian Magnetic Resonance Imaging with R2D2

Yiwei Chen, Chao Tang, Amir Aghabiglou, Chung San Chu, Yves Wiaux

We propose a new approach for non-Cartesian magnetic resonance image reconstruction. While unrolled architectures provide robustness via data-consistency layers, embedding measurement operators in Deep Neural Network (DNN) can become impractical at large scale. Alternative Plug-and-Play (PnP) approaches, where the denoising DNNs are blind to the measurement setting, are not affected by this limitation and have also proven effective, but their highly iterative nature also affects scalability. To address this scalability challenge, we leverage the "Residual-to-Residual DNN series for high-Dynamic range imaging (R2D2)" approach recently introduced in astronomical imaging. R2D2's reconstruction is formed as a series of residual images, iteratively estimated as outputs of DNNs taking the previous iteration's image estimate and associated data residual as inputs. The method can be interpreted as a learned version of the Matching Pursuit algorithm. We demonstrate R2D2 in simulation, considering radial k-space sampling acquisition sequences. Our preliminary results suggest that R2D2 achieves: (i) suboptimal performance compared to its unrolled incarnation R2D2-Net, which is however non-scalable due to the necessary embedding of NUFFT-based data-consistency layers; (ii) superior reconstruction quality to a scalable version of R2D2-Net embedding an FFT-based approximation for data consistency; (iii) superior reconstruction quality to PnP, while only requiring few iterations.

comment: submitted to IEEE EUSIPCO 2024

♻ ☆ Asymptotic Bayes risk of semi-supervised learning with uncertain labeling

Victor Leger, Romain Couillet

This article considers a semi-supervised classification setting on a Gaussian mixture model, where the data is not labeled strictly as usual, but instead with uncertain labels. Our main aim is to compute the Bayes risk for this model. We compare the behavior of the Bayes risk and the best known algorithm for this model. This comparison eventually gives new insights over the algorithm.

♻ ☆ Coarse-Tuning for Ad-hoc Document Retrieval Using Pre-trained Language Models LREC

Atsushi Keyaki, Ribeka Keyaki

Fine-tuning in information retrieval systems using pre-trained language models (PLM-based IR) requires learning query representations and query-document relations, in addition to downstream task-specific learning. This study introduces coarse-tuning as an intermediate learning stage that bridges pre-training and fine-tuning. By learning query representations and query-document relations in coarse-tuning, we aim to reduce the load of fine-tuning and improve the learning effect of downstream IR tasks. We propose Query-Document Pair Prediction (QDPP) for coarse-tuning, which predicts the appropriateness of query-document pairs. Evaluation experiments show that the proposed method significantly improves MRR and/or nDCG@5 in four ad-hoc document retrieval datasets. Furthermore, the results of the query prediction task suggested that coarse-tuning facilitated learning of query representation and query-document relations.

comment: Accepted at LREC-COLING 2024

♻ ☆ Expectations Versus Reality: Evaluating Intrusion Detection Systems in Practice

Jake Hesford, Daniel Cheng, Alan Wan, Larry Huynh, Seungho Kim, Hyoungshick Kim, Jin B. Hong

Our paper provides empirical comparisons between recent IDSs to provide an objective comparison between them to help users choose the most appropriate solution based on their requirements. Our results show that no one solution is the best, but is dependent on external variables such as the types of attacks, complexity, and network environment in the dataset. For example, BoT_IoT and Stratosphere IoT datasets both capture IoT-related attacks, but the deep neural network performed the best when tested using the BoT_IoT dataset while HELAD performed the best when tested using the Stratosphere IoT dataset. So although we found that a deep neural network solution had the highest average F1 scores on tested datasets, it is not always the best-performing one. We further discuss difficulties in using IDS from literature and project repositories, which complicated drawing definitive conclusions regarding IDS selection.

comment: 10 pages

♻ ☆ Imitating Cost-Constrained Behaviors in Reinforcement Learning

Qian Shao, Pradeep Varakantham, Shih-Fen Cheng

Complex planning and scheduling problems have long been solved using various optimization or heuristic approaches. In recent years, imitation learning that aims to learn from expert demonstrations has been proposed as a viable alternative to solving these problems. Generally speaking, imitation learning is designed to learn either the reward (or preference) model or directly the behavioral policy by observing the behavior of an expert. Existing work in imitation learning and inverse reinforcement learning has focused on imitation primarily in unconstrained settings (e.g., no limit on fuel consumed by the vehicle). However, in many real-world domains, the behavior of an expert is governed not only by reward (or preference) but also by constraints. For instance, decisions on self-driving delivery vehicles are dependent not only on the route preferences/rewards (depending on past demand data) but also on the fuel in the vehicle and the time available. In such problems, imitation learning is challenging as decisions are not only dictated by the reward model but are also dependent on a cost-constrained model. In this paper, we provide multiple methods that match expert distributions in the presence of trajectory cost constraints through (a) Lagrangian-based method; (b) Meta-gradients to find a good trade-off between expected return and minimizing constraint violation; and (c) Cost-violation-based alternating gradient. We empirically show that leading imitation learning approaches imitate cost-constrained behaviors poorly and our meta-gradient-based approach achieves the best performance.

comment: Accepted to the 34th International Conference on Automated Planning and Scheduling (ICAPS-24)

♻ ☆ Language Models are Free Boosters for Biomedical Imaging Tasks

Zhixin Lai, Jing Wu, Suiyao Chen, Yucheng Zhou, Naira Hovakimyan

In this study, we uncover the unexpected efficacy of residual-based large language models (LLMs) as part of encoders for biomedical imaging tasks, a domain traditionally devoid of language or textual data. The approach diverges from established methodologies by utilizing a frozen transformer block, extracted from pre-trained LLMs, as an innovative encoder layer for the direct processing of visual tokens. This strategy represents a significant departure from the standard multi-modal vision-language frameworks, which typically hinge on language-driven prompts and inputs. We found that these LLMs could boost performance across a spectrum of biomedical imaging applications, including both 2D and 3D visual classification tasks, serving as plug-and-play boosters. More interestingly, as a byproduct, we found that the proposed framework achieved superior performance, setting new state-of-the-art results on extensive, standardized datasets in MedMNIST-2D and 3D. Through this work, we aim to open new avenues for employing LLMs in biomedical imaging and enriching the understanding of their potential in this specialized domain.

♻ ☆ World Models via Policy-Guided Trajectory Diffusion

Marc Rigter, Jun Yamada, Ingmar Posner

World models are a powerful tool for developing intelligent agents. By predicting the outcome of a sequence of actions, world models enable policies to be optimised via on-policy reinforcement learning (RL) using synthetic data, i.e. in "in imagination". Existing world models are autoregressive in that they interleave predicting the next state with sampling the next action from the policy. Prediction error inevitably compounds as the trajectory length grows. In this work, we propose a novel world modelling approach that is not autoregressive and generates entire on-policy trajectories in a single pass through a diffusion model. Our approach, Policy-Guided Trajectory Diffusion (PolyGRAD), leverages a denoising model in addition to the gradient of the action distribution of the policy to diffuse a trajectory of initially random states and actions into an on-policy synthetic trajectory. We analyse the connections between PolyGRAD, score-based generative models, and classifier-guided diffusion models. Our results demonstrate that PolyGRAD outperforms state-of-the-art baselines in terms of trajectory prediction error for short trajectories, with the exception of autoregressive diffusion. For short trajectories, PolyGRAD obtains similar errors to autoregressive diffusion, but with lower computational requirements. For long trajectories, PolyGRAD obtains comparable performance to baselines. Our experiments demonstrate that PolyGRAD enables performant policies to be trained via on-policy RL in imagination for MuJoCo continuous control domains. Thus, PolyGRAD introduces a new paradigm for accurate on-policy world modelling without autoregressive sampling.

comment: Published in TMLR, March 2024

♻ ☆ A Physics-embedded Deep Learning Framework for Cloth Simulation

Zhiwei Zhao

Delicate cloth simulations have long been desired in computer graphics. Various methods were proposed to improve engaged force interactions, collision handling, and numerical integrations. Deep learning has the potential to achieve fast and real-time simulation, but common neural network structures often demand many parameters to capture cloth dynamics. This paper proposes a physics-embedded learning framework that directly encodes physical features of cloth simulation. The convolutional neural network is used to represent spatial correlations of the mass-spring system, after which three branches are designed to learn linear, nonlinear, and time derivate features of cloth physics. The framework can also integrate with other external forces and collision handling through either traditional simulators or sub neural networks. The model is tested across different cloth animation cases, without training with new data. Agreement with baselines and predictive realism successfully validate its generalization ability. Inference efficiency of the proposed model also defeats traditional physics simulation. This framework is also designed to easily integrate with other visual refinement techniques like wrinkle carving, which leaves significant chances to incorporate prevailing macing learning techniques in 3D cloth amination.

comment: A derivation is incomplete, and updations are being processed

☆ SLEDGE: Synthesizing Simulation Environments for Driving Agents with Generative Models

Kashyap Chitta, Daniel Dauner, Andreas Geiger

SLEDGE is the first generative simulator for vehicle motion planning trained on real-world driving logs. Its core component is a learned model that is able to generate agent bounding boxes and lane graphs. The model's outputs serve as an initial state for traffic simulation. The unique properties of the entities to be generated for SLEDGE, such as their connectivity and variable count per scene, render the naive application of most modern generative models to this task non-trivial. Therefore, together with a systematic study of existing lane graph representations, we introduce a novel raster-to-vector autoencoder (RVAE). It encodes agents and the lane graph into distinct channels in a rasterized latent map. This facilitates both lane-conditioned agent generation and combined generation of lanes and agents with a Diffusion Transformer. Using generated entities in SLEDGE enables greater control over the simulation, e.g. upsampling turns or increasing traffic density. Further, SLEDGE can support 500m long routes, a capability not found in existing data-driven simulators like nuPlan. It presents new challenges for planning algorithms, evidenced by failure rates of over 40% for PDM, the winner of the 2023 nuPlan challenge, when tested on hard routes and dense traffic generated by our model. Compared to nuPlan, SLEDGE requires 500$\times$ less storage to set up (<4GB), making it a more accessible option and helping with democratizing future research in this field.

☆ Multi-Agent Clarity-Aware Dynamic Coverage with Gaussian Processes

Devansh R. Agrawal, Dimitra Panagou

This paper presents two algorithms for multi-agent dynamic coverage in spatiotemporal environments, where the coverage algorithms are informed by the method of data assimilation. In particular, we show that by considering the information assimilation algorithm, here a Numerical Gaussian Process Kalman Filter, the influence of measurements taken at one position on the uncertainty of the estimate at another location can be computed. We use this relationship to propose new coverage algorithms. Furthermore, we show that the controllers naturally extend to the multi-agent context, allowing for a distributed-control central-information paradigm for multi-agent coverage. Finally, we demonstrate the algorithms through a realistic simulation of a team of UAVs collecting wind data over a region in Austria.

comment: 8 pages, 2 figures, submitted to CDC 2024

☆ CMP: Cooperative Motion Prediction with Multi-Agent Communication

Zhuoyuan Wu, Yuping Wang, Hengbo Ma, Zhaowei Li, Hang Qiu, Jiachen Li

The confluence of the advancement of Autonomous Vehicles (AVs) and the maturity of Vehicle-to-Everything (V2X) communication has enabled the capability of cooperative connected and automated vehicles (CAVs). Building on top of cooperative perception, this paper explores the feasibility and effectiveness of cooperative motion prediction. Our method, CMP, takes LiDAR signals as input to enhance tracking and prediction capabilities. Unlike previous work that focuses separately on either cooperative perception or motion prediction, our framework, to the best of our knowledge, is the first to address the unified problem where CAVs share information in both perception and prediction modules. Incorporated into our design is the unique capability to tolerate realistic V2X bandwidth limitations and transmission delays, while dealing with bulky perception representations. We also propose a prediction aggregation module, which unifies the predictions obtained by different CAVs and generates the final prediction. Through extensive experiments and ablation studies, we demonstrate the effectiveness of our method in cooperative perception, tracking, and motion prediction tasks. In particular, CMP reduces the average prediction error by 17.2\% with fewer missing detections compared with the no cooperation setting. Our work marks a significant step forward in the cooperative capabilities of CAVs, showcasing enhanced performance in complex scenarios.

☆ Multi Agent Pathfinding for Noise Restricted Hybrid Fuel Unmanned Aerial Vehicles

Drew Scott, Satyanarayana G. Manyam, David W. Casbeer, Manish Kumar, Isaac E. Weintraub

Multi Agent Path Finding (MAPF) seeks the optimal set of paths for multiple agents from respective start to goal locations such that no paths conflict. We address the MAPF problem for a fleet of hybrid-fuel unmanned aerial vehicles which are subject to location-dependent noise restrictions. We solve this problem by searching a constraint tree for which the subproblem at each node is a set of shortest path problems subject to the noise and fuel constraints and conflict zone avoidance. A labeling algorithm is presented to solve this subproblem, including the conflict zones which are treated as dynamic obstacles. We present the experimental results of the algorithms for various graph sizes and number of agents.

comment: 6 pages, 7 figures

★ Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation

Abdelrhman Werby, Chenguang Huang, Martin Büchner, Abhinav Valada, Wolfram Burgard

Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features. While these maps allow for the prediction of point-wise saliency maps when queried for a certain language concept, large-scale environments and abstract queries beyond the object level still pose a considerable hurdle, ultimately limiting language-grounded robotic navigation. In this work, we present HOV-SG, a hierarchical open-vocabulary 3D scene graph mapping approach for language-grounded robot navigation. Leveraging open-vocabulary vision foundation models, we first obtain state-of-the-art open-vocabulary segment-level maps in 3D and subsequently construct a 3D scene graph hierarchy consisting of floor, room, and object concepts, each enriched with open-vocabulary features. Our approach is able to represent multi-story buildings and allows robotic traversal of those using a cross-floor Voronoi graph. HOV-SG is evaluated on three distinct datasets and surpasses previous baselines in open-vocabulary semantic accuracy on the object, room, and floor level while producing a 75% reduction in representation size compared to dense open-vocabulary maps. In order to prove the efficacy and generalization capabilities of HOV-SG, we showcase successful long-horizon language-conditioned robot navigation within real-world multi-storage environments. We provide code and trial video data at http://hovsg.github.io/.

comment: Code and video are available at http://hovsg.github.io/

☆ Scenario-Based Curriculum Generation for Multi-Agent Autonomous Driving

Axel Brunnbauer, Luigi Berducci, Peter Priller, Dejan Nickovic, Radu Grosu

The automated generation of diverse and complex training scenarios has been an important ingredient in many complex learning tasks. Especially in real-world application domains, such as autonomous driving, auto-curriculum generation is considered vital for obtaining robust and general policies. However, crafting traffic scenarios with multiple, heterogeneous agents is typically considered as a tedious and time-consuming task, especially in more complex simulation environments. In our work, we introduce MATS-Gym, a Multi-Agent Traffic Scenario framework to train agents in CARLA, a high-fidelity driving simulator. MATS-Gym is a multi-agent training framework for autonomous driving that uses partial scenario specifications to generate traffic scenarios with variable numbers of agents. This paper unifies various existing approaches to traffic scenario description into a single training framework and demonstrates how it can be integrated with techniques from unsupervised environment design to automate the generation of adaptive auto-curricula. The code is available at https://github.com/AutonomousDrivingExaminer/mats-gym.

comment: 7 Pages, Under Review

☆ System Calibration of a Field Phenotyping Robot with Multiple High-Precision Profile Laser Scanners

Felix Esser, Gereon Tombrink, Andre Cornelißen, Lasse Klingbeil, Heiner Kuhlmann

The creation of precise and high-resolution crop point clouds in agricultural fields has become a key challenge for high-throughput phenotyping applications. This work implements a novel calibration method to calibrate the laser scanning system of an agricultural field robot consisting of two industrial-grade laser scanners used for high-precise 3D crop point cloud creation. The calibration method optimizes the transformation between the scanner origins and the robot pose by minimizing 3D point omnivariances within the point cloud. Moreover, we present a novel factor graph-based pose estimation method that fuses total station prism measurements with IMU and GNSS heading information for high-precise pose determination during calibration. The root-mean-square error of the distances to a georeferenced ground truth point cloud results in 0.8 cm after parameter optimization. Furthermore, our results show the importance of a reference point cloud in the calibration method needed to estimate the vertical translation of the calibration. Challenges arise due to non-static parameters while the robot moves, indicated by systematic deviations to a ground truth terrestrial laser scan.

☆ Optical Flow Based Detection and Tracking of Moving Objects for Autonomous Vehicles TITS

MReza Alipour Sormoli, Mehrdad Dianati, Sajjad Mozaffari, Roger woodman

Accurate velocity estimation of surrounding moving objects and their trajectories are critical elements of perception systems in Automated/Autonomous Vehicles (AVs) with a direct impact on their safety. These are non-trivial problems due to the diverse types and sizes of such objects and their dynamic and random behaviour. Recent point cloud based solutions often use Iterative Closest Point (ICP) techniques, which are known to have certain limitations. For example, their computational costs are high due to their iterative nature, and their estimation error often deteriorates as the relative velocities of the target objects increase (>2 m/sec). Motivated by such shortcomings, this paper first proposes a novel Detection and Tracking of Moving Objects (DATMO) for AVs based on an optical flow technique, which is proven to be computationally efficient and highly accurate for such problems. \textcolor{black}{This is achieved by representing the driving scenario as a vector field and applying vector calculus theories to ensure spatiotemporal continuity.} We also report the results of a comprehensive performance evaluation of the proposed DATMO technique, carried out in this study using synthetic and real-world data. The results of this study demonstrate the superiority of the proposed technique, compared to the DATMO techniques in the literature, in terms of estimation accuracy and processing time in a wide range of relative velocities of moving objects. Finally, we evaluate and discuss the sensitivity of the estimation error of the proposed DATMO technique to various system and environmental parameters, as well as the relative velocities of the moving objects.

comment: This manuscript has been accepted as a regular paper in Transactions on Intelligent Transportation Systems (DOI: 10.1109/TITS.2024.3382495)

☆ LiDAR-Based Crop Row Detection Algorithm for Over-Canopy Autonomous Navigation in Agriculture Fields IROS 2024

Ruiji Liu, Francisco Yandun, George Kantor

Autonomous navigation is crucial for various robotics applications in agriculture. However, many existing methods depend on RTK-GPS systems, which are expensive and susceptible to poor signal coverage. This paper introduces a state-of-the-art LiDAR-based navigation system that can achieve over-canopy autonomous navigation in row-crop fields, even when the canopy fully blocks the interrow spacing. Our crop row detection algorithm can detect crop rows across diverse scenarios, encompassing various crop types, growth stages, weed presence, and discontinuities within the crop rows. Without utilizing the global localization of the robot, our navigation system can perform autonomous navigation in these challenging scenarios, detect the end of the crop rows, and navigate to the next crop row autonomously, providing a crop-agnostic approach to navigate the whole row-crop field. This navigation system has undergone tests in various simulated agricultural fields, achieving an average of $2.98cm$ autonomous driving accuracy without human intervention on the custom Amiga robot. In addition, the qualitative results of our crop row detection algorithm from the actual soybean fields validate our LiDAR-based crop row detection algorithm's potential for practical agricultural applications.

comment: 7 pages, 9 figures, submitted to IROS 2024

☆ Learning Goal-Directed Object Pushing in Cluttered Scenes with Location-Based Attention IROS

Nils Dengler, Juan Del Aguila Ferrandis, João Moura, Sethu Vijayakumar, Maren Bennewitz

Non-prehensile planar pushing is a challenging task due to its underactuated nature with hybrid-dynamics, where a robot needs to reason about an object's long-term behaviour and contact-switching, while being robust to contact uncertainty. The presence of clutter in the environment further complicates this task, introducing the need to include more sophisticated spatial analysis to avoid collisions. Building upon prior work on reinforcement learning (RL) with multimodal categorical exploration for planar pushing, in this paper we incorporate location-based attention to enable robust navigation through clutter. Unlike previous RL literature addressing this obstacle avoidance pushing task, our framework requires no predefined global paths and considers the target orientation of the manipulated object. Our results demonstrate that the learned policies successfully navigate through a wide range of complex obstacle configurations, including dynamic obstacles, with smooth motions, achieving the desired target object pose. We also validate the transferability of the learned policies to robotic hardware using the KUKA iiwa robot arm.

comment: Submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

☆ UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection with Sparse LiDAR and Large Domain Gaps

Maciej K Wozniak, Mattias Hansson, Marko Thiel, Patric Jensfelt

In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains. Our code is open-source and will be available soon.

☆ Online Tree Reconstruction and Forest Inventory on a Mobile Robotic System

Leonard Freißmuth, Matias Mattamala, Nived Chebrolu, Simon Schaefer, Stefan Leutenegger, Maurice Fallon

Terrestrial laser scanning (TLS) is the standard technique used to create accurate point clouds for digital forest inventories. However, the measurement process is demanding, requiring up to two days per hectare for data collection, significant data storage, as well as resource-heavy post-processing of 3D data. In this work, we present a real-time mapping and analysis system that enables online generation of forest inventories using mobile laser scanners that can be mounted e.g. on mobile robots. Given incrementally created and locally accurate submaps-data payloads-our approach extracts tree candidates using a custom, Voronoi-inspired clustering algorithm. Tree candidates are reconstructed using an adapted Hough algorithm, which enables robust modeling of the tree stem. Further, we explicitly incorporate the incremental nature of the data collection by consistently updating the database using a pose graph LiDAR SLAM system. This enables us to refine our estimates of the tree traits if an area is revisited later during a mission. We demonstrate competitive accuracy to TLS or manual measurements using laser scanners that we mounted on backpacks or mobile robots operating in conifer, broad-leaf and mixed forests. Our results achieve RMSE of 1.93 cm, a bias of 0.65 cm and a standard deviation of 1.81 cm (averaged across these sequences)-with no post-processing required after the mission is complete.

☆ Interactive Identification of Granular Materials using Force Measurements IROS 2024

Samuli Hynninen, Tran Nguyen Le, Ville Kyrki

The ability to identify granular materials facilitates the emergence of various new applications in robotics, ranging from cooking at home to truck loading at mining sites. However, granular material identification remains a challenging and underexplored area. In this work, we present a novel interactive material identification framework that enables robots to identify a wide range of granular materials using only a force-torque sensor for perception. Our framework, comprising interactive exploration, feature extraction, and classification stages, prioritizes simplicity and transparency for seamless integration into various manipulation pipelines. We evaluate the proposed approach through extensive experiments with a real-world dataset comprising 11 granular materials, which we also make publicly available. Additionally, we conducted a comprehensive qualitative analysis of the dataset to offer deeper insights into its nature, aiding future development. Our results show that the proposed method is capable of accurately identifying a wide range of granular materials solely relying on force measurements obtained from direct interaction with the materials. Code and dataset are available at: https://irobotics.aalto.fi/indentify_granular/.

comment: Submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024)

☆ Aerial Robots Carrying Flexible Cables: Dynamic Shape Optimal Control via Spectral Method Model

Yaolei Shen, Chiara Gabellieri, Antonio Franchi

In this work, we present a model-based optimal boundary control design for an aerial robotic system composed of a quadrotor carrying a flexible cable. The whole system is modeled by partial differential equations (PDEs) combined with boundary conditions described by ordinary differential equations (ODEs). The proper orthogonal decomposition (POD) method is adopted to project the original infinite-dimensional system on a subspace spanned by orthogonal basis functions. Based on the reduced order model, nonlinear model predictive control (NMPC) is implemented online to realize shape trajectory tracking of the flexible cable in an optimal predictive fashion. The proposed reduced modeling and optimal control paradigms are numerically verified against an accurate high-dimensional FDM-based model in different scenarios and the controller's superior performance is shown compared to an optimally tuned PID controller.

★ Time-Optimal Flight with Safety Constraints and Data-driven Dynamics

Maria Krinner, Angel Romero, Leonard Bauersfeld, Melanie Zeilinger, Andrea Carron, Davide Scaramuzza

Time-optimal quadrotor flight is an extremely challenging problem due to the limited control authority encountered at the limit of handling. Model Predictive Contouring Control (MPCC) has emerged as a leading model-based approach for time optimization problems such as drone racing. However, the standard MPCC formulation used in quadrotor racing introduces the notion of the gates directly in the cost function, creating a multi-objective optimization that continuously trades off between maximizing progress and tracking the path accurately. This paper introduces three key components that enhance the MPCC approach for drone racing. First and foremost, we provide safety guarantees in the form of a constraint and terminal set. The safety set is designed as a spatial constraint which prevents gate collisions while allowing for time-optimization only in the cost function. Second, we augment the existing first principles dynamics with a residual term that captures complex aerodynamic effects and thrust forces learned directly from real world data. Third, we use Trust Region Bayesian Optimization (TuRBO), a state of the art global Bayesian Optimization algorithm, to tune the hyperparameters of the MPC controller given a sparse reward based on lap time minimization. The proposed approach achieves similar lap times to the best state-of-the-art RL and outperforms the best time-optimal controller while satisfying constraints. In both simulation and real-world, our approach consistently prevents gate crashes with 100\% success rate, while pushing the quadrotor to its physical limit reaching speeds of more than 80km/h.

comment: 12 pages, 7 figures

☆ DeepMIF: Deep Monotonic Implicit Fields for Large-Scale LiDAR 3D Mapping

Kutay Yılmaz, Matthias Nießner, Anastasiia Kornilova, Alexey Artemov

Recently, significant progress has been achieved in sensing real large-scale outdoor 3D environments, particularly by using modern acquisition equipment such as LiDAR sensors. Unfortunately, they are fundamentally limited in their ability to produce dense, complete 3D scenes. To address this issue, recent learning-based methods integrate neural implicit representations and optimizable feature grids to approximate surfaces of 3D scenes. However, naively fitting samples along raw LiDAR rays leads to noisy 3D mapping results due to the nature of sparse, conflicting LiDAR measurements. Instead, in this work we depart from fitting LiDAR data exactly, instead letting the network optimize a non-metric monotonic implicit field defined in 3D space. To fit our field, we design a learning system integrating a monotonicity loss that enables optimizing neural monotonic fields and leverages recent progress in large-scale 3D mapping. Our algorithm achieves high-quality dense 3D mapping performance as captured by multiple quantitative and perceptual measures and visual results obtained for Mai City, Newer College, and KITTI benchmarks. The code of our approach will be made publicly available.

comment: 8 pages, 6 figures

☆ Design and Preliminary Evaluation of a Torso Stabiliser for Individuals with Spinal Cord Injury

Rejin John Varghese, Man-Yan Tong, Isabella Szczech, Peter Bryan, Dario Farina, Etienne Burdet

Spinal cord injuries (SCIs) generally result in sensory and mobility impairments, with torso instability being particularly debilitating. Existing torso stabilisers are often rigid and restrictive. This paper presents an early investigation into a non-restrictive 1 degree-of-freedom (DoF) mechanical torso stabiliser inspired by devices such as centrifugal clutches and seat-belt mechanisms. Firstly, the paper presents a motion-capture (MoCap) and OpenSim-based kinematic analysis of the cable-based system to understand requisite device characteristics. The simulated evaluation resulted in the cable-based device to require 55-60cm of unrestricted travel, and to lock at a threshold cable velocity of 80-100cm/sec. Next, the developed 1-DoF device is introduced. The proposed mechanical device is transparent during activities of daily living, and transitions to compliant blocking when incipient fall is detected. Prototype behaviour was then validated using a MoCap-based kinematic analysis to verify non-restrictive movement, reliable transition to blocking, and compliance of the blocking.

comment: 4 pages, 4 figures, 10 references. Submitted to IEEE EMBC 2024 conference

☆ High-Power, Flexible, Robust Hand: Development of Musculoskeletal Hand Using Machined Springs and Realization of Self-Weight Supporting Motion with Humanoid IROS2017

Shogo Makino, Kento Kawaharazuka, Masaya Kawamura, Yuki Asano, Kei Okada, Masayuki Inaba

Human can not only support their body during standing or walking, but also support them by hand, so that they can dangle a bar and others. But most humanoid robots support their body only in the foot and they use their hand just to manipulate objects because their hands are too weak to support their body. Strong hands are supposed to enable humanoid robots to act in much broader scene. Therefore, we developed new life-size five-fingered hand that can support the body of life-size humanoid robot. It is tendon-driven and underactuated hand and actuators in forearms produce large gripping force. This hand has flexible joints using machined springs, which can be designed integrally with the attachment. Thus, it has both structural strength and impact resistance in spite of small size. As other characteristics, this hand has force sensors to measure external force and the fingers can be flexed along objects though the number of actuators to flex fingers is less than that of fingers. We installed the developed hand on musculoskeletal humanoid "Kengoro" and achieved two self-weight supporting motions: push-up motion and dangling motion.

comment: accepted at IROS2017

☆ Five-fingered Hand with Wide Range of Thumb Using Combination of Machined Springs and Variable Stiffness Joints IROS2018

Shogo Makino, Kento Kawaharazuka, Ayaka Fujii, Masaya Kawamura, Tasuku Makabe, Moritaka Onitsuka, Yuki Asano, Kei Okada, Koji Kawasaki, Masayuki Inaba

Human hands can not only grasp objects of various shape and size and manipulate them in hands but also exert such a large gripping force that they can support the body in the situations such as dangling a bar and climbing a ladder. On the other hand, it is difficult for most robot hands to manage both. Therefore in this paper we developed the hand which can grasp various objects and exert large gripping force. To develop such hand, we focused on the thumb CM joint with wide range of motion and the MP joints of four fingers with the DOF of abduction and adduction. Based on the hand with large gripping force and flexibility using machined spring, we applied above mentioned joint mechanism to the hand. The thumb CM joint has wide range of motion because of the combination of three machined springs and MP joints of four fingers have variable rigidity mechanism instead of driving each joint independently in order to move joint in limited space and by limited actuators. Using the developed hand, we achieved the grasping of various objects, supporting a large load and several motions with an arm.

comment: accepted at IROS2018

☆ Adaptive Line-Of-Sight guidance law based on vector fields path following for underactuated unmanned surface vehicle

Jie Qi, Ronghua Wang, Nailong Wu, Yuxin Fan, Jigang Wang

The focus of this paper is to develop a methodology that enables an unmanned surface vehicle (USV) to efficiently track a planned path. The introduction of a vector field-based adaptive line-of-sight guidance law (VFALOS) for accurate trajectory tracking and minimizing the overshoot response time during USV tracking of curved paths improves the overall line-of-sight (LOS) guidance method. These improvements contribute to faster convergence to the desired path, reduce oscillations, and can mitigate the effects of persistent external disturbances. It is shown that the proposed guidance law exhibits k-exponential stability when converging to the desired path consisting of straight and curved lines. The results in the paper show that the proposed method effectively improves the accuracy of the USV tracking the desired path while ensuring the safety of the USV work.

☆ Adaptive LiDAR-Radar Fusion for Outdoor Odometry Across Dense Smoke Conditions

Chiyun Noh, Ayoung Kim

Robust odometry estimation in perceptually degraded environments represents a key challenge in the field of robotics. In this paper, we propose a LiDAR-radar fusion method for robust odometry for adverse environment with LiDAR degeneracy. By comparing the LiDAR point cloud with the radar static point cloud obtained through preprocessing module, it is possible to identify instances of LiDAR degeneracy to overcome perceptual limits. We demonstrate the effectiveness of our method in challenging conditions such as dense smoke, showcasing its ability to reliably estimate odometry and identify/remove dynamic points prone to LiDAR degeneracy.

☆ Cyclic pursuit formation control for arbitrary desired shapes

Anna Fujioka, Masaki Ogura, Naoki Wakamiya

A multi-agent system comprises numerous agents that autonomously make decisions to collectively accomplish tasks, drawing significant attention for their wide-ranging applications. Within this context, formation control emerges as a prominent task, wherein agents collaboratively shape and maneuver while preserving formation integrity. Our focus centers on cyclic pursuit, a method facilitating the formation of circles, ellipses, and figure-eights under the assumption that agents can only perceive the relative positions of those preceding them. However, this method's scope has been restricted to these specific shapes, leaving the feasibility of forming other shapes uncertain. In response, our study proposes a novel method based on cyclic pursuit capable of forming a broader array of shapes, enabling agents to individually shape while pursuing preceding agents, thereby extending the repertoire of achievable formations. We present two scenarios concerning the information available to agents and devise formation control methods tailored to each scenario. Through extensive simulations, we demonstrate the efficacy of our proposed method in forming multiple shapes, including those represented as Fourier series, thereby underscoring the versatility and effectiveness of our approach.

☆ Natural-artificial hybrid swarm: Cyborg-insect group navigation in unknown obstructed soft terrain

Yang Bai, Phuoc Thanh Tran Ngoc, Huu Duoc Nguyen, Duc Long Le, Quang Huy Ha, Kazuki Kai, Yu Xiang See To, Yaosheng Deng, Jie Song, Naoki Wakamiya, Hirotaka Sato, Masaki Ogura

Navigating multi-robot systems in complex terrains has always been a challenging task. This is due to the inherent limitations of traditional robots in collision avoidance, adaptation to unknown environments, and sustained energy efficiency. In order to overcome these limitations, this research proposes a solution by integrating living insects with miniature electronic controllers to enable robotic-like programmable control, and proposing a novel control algorithm for swarming. Although these creatures, called cyborg insects, have the ability to instinctively avoid collisions with neighbors and obstacles while adapting to complex terrains, there is a lack of literature on the control of multi-cyborg systems. This research gap is due to the difficulty in coordinating the movements of a cyborg system under the presence of insects' inherent individual variability in their reactions to control input. In response to this issue, we propose a novel swarm navigation algorithm addressing these challenges. The effectiveness of the algorithm is demonstrated through an experimental validation in which a cyborg swarm was successfully navigated through an unknown sandy field with obstacles and hills. This research contributes to the domain of swarm robotics and showcases the potential of integrating biological organisms with robotics and control theory to create more intelligent autonomous systems with real-world applications.

☆ Multi-Objective Trajectory Planning with Dual-Encoder

Beibei Zhang, Tian Xiang, Chentao Mao, Yuhua Zheng, Shuai Li, Haoyi Niu, Xiangming Xi, Wenyuan Bai, Feng Gao

Time-jerk optimal trajectory planning is crucial in advancing robotic arms' performance in dynamic tasks. Traditional methods rely on solving complex nonlinear programming problems, bringing significant delays in generating optimized trajectories. In this paper, we propose a two-stage approach to accelerate time-jerk optimal trajectory planning. Firstly, we introduce a dual-encoder based transformer model to establish a good preliminary trajectory. This trajectory is subsequently refined through sequential quadratic programming to improve its optimality and robustness. Our approach outperforms the state-of-the-art by up to 79.72\% in reducing trajectory planning time. Compared with existing methods, our method shrinks the optimality gap with the objective function value decreasing by up to 29.9\%.

comment: 6 pages, 7 figures, conference

☆ Unified Path and Gait Planning for Safe Bipedal Robot Navigation

Chengyang Peng, Victor Paredes, Ayonga Hereid

Safe path and gait planning are essential for bipedal robots to navigate complex real-world environments. The prevailing approaches often plan the path and gait separately in a hierarchical fashion, potentially resulting in unsafe movements due to neglecting the physical constraints of walking robots. A safety-critical path must not only avoid obstacles but also ensure that the robot's gaits are subject to its dynamic and kinematic constraints. This work presents a novel approach that unifies path planning and gait planning via a Model Predictive Control (MPC) using the Linear Inverted Pendulum (LIP) model representing bipedal locomotion. This approach considers environmental constraints, such as obstacles, and the robot's kinematics and dynamics constraints. By using discrete-time Control Barrier Functions for obstacle avoidance, our approach generates the next foot landing position, ensuring robust walking gaits and a safe navigation path within clustered environments. We validated our proposed approach in simulation using a Digit robot in 20 randomly created environments. The results demonstrate improved performance in terms of safety and robustness when compared to hierarchical path and gait planning frameworks.

comment: 8 pages

★ Sparse-Graph-Enabled Formation Planning for Large-Scale Aerial Swarms

Yuan Zhou, Lun Quan, Chao Xu, Guangtong Xu, Fei Gao

The formation trajectory planning using complete graphs to model collaborative constraints becomes computationally intractable as the number of drones increases due to the curse of dimensionality. To tackle this issue, this paper presents a sparse graph construction method for formation planning to realize better efficiency-performance trade-off. Firstly, a sparsification mechanism for complete graphs is designed to ensure the global rigidity of sparsified graphs, which is a necessary condition for uniquely corresponding to a geometric shape. Secondly, a good sparse graph is constructed to preserve the main structural feature of complete graphs sufficiently. Since the graph-based formation constraint is described by Laplacian matrix, the sparse graph construction problem is equivalent to submatrix selection, which has combinatorial time complexity and needs a scoring metric. Via comparative simulations, the Max-Trace matrix-revealing metric shows the promising performance. The sparse graph is integrated into the formation planning. Simulation results with 72 drones in complex environments demonstrate that when preserving 30\% connection edges, our method has comparative formation error and recovery performance w.r.t. complete graphs. Meanwhile, the planning efficiency is improved by approximate an order of magnitude. Benchmark comparisons and ablation studies are conducted to fully validate the merits of our method.

♻ ☆ Towards Massive Interaction with Generalist Robotics: A Systematic Review of XR-enabled Remote Human-Robot Interaction Systems

Xian Wang, Luyao Shen, Lik-Hang Lee

The rising interest of generalist robots seek to create robots with versatility to handle multiple tasks in a variety of environments, and human will interact with such robots through immersive interfaces. In the context of human-robot interaction (HRI), this survey provides an exhaustive review of the applications of extended reality (XR) technologies in the field of remote HRI. We developed a systematic search strategy based on the PRISMA methodology. From the initial 2,561 articles selected, 100 research papers that met our inclusion criteria were included. We categorized and summarized the domain in detail, delving into XR technologies, including augmented reality (AR), virtual reality (VR), and mixed reality (MR), and their applications in facilitating intuitive and effective remote control and interaction with robotic systems. The survey highlights existing articles on the application of XR technologies, user experience enhancement, and various interaction designs for XR in remote HRI, providing insights into current trends and future directions. We also identified potential gaps and opportunities for future research to improve remote HRI systems through XR technology to guide and inform future XR and robotics research.

♻ ☆ Guessing human intentions to avoid dangerous situations in caregiving robots IROS2024

Noé Zapata, Gerardo Pérez, Lucas Bonilla, Pedro Núñez, Pilar Bachiller, Pablo Bustos

For robots to interact socially, they must interpret human intentions and anticipate their potential outcomes accurately. This is particularly important for social robots designed for human care, which may face potentially dangerous situations for people, such as unseen obstacles in their way, that should be avoided. This paper explores the Artificial Theory of Mind (ATM) approach to inferring and interpreting human intentions. We propose an algorithm that detects risky situations for humans, selecting a robot action that removes the danger in real time. We use the simulation-based approach to ATM and adopt the 'like-me' policy to assign intentions and actions to people. Using this strategy, the robot can detect and act with a high rate of success under time-constrained situations. The algorithm has been implemented as part of an existing robotics cognitive architecture and tested in simulation scenarios. Three experiments have been conducted to test the implementation's robustness, precision and real-time response, including a simulated scenario, a human-in-the-loop hybrid configuration and a real-world scenario.

comment: 8 pages, 6 figures. Submitted to IROS2024. For associated mpeg file see https://youtu.be/87UEB8P97KY

♻ ☆ Safe Explicable Planning

Akkamahadevi Hanni, Andrew Boateng, Yu Zhang

Human expectations arise from their understanding of others and the world. In the context of human-AI interaction, this understanding may not align with reality, leading to the AI agent failing to meet expectations and compromising team performance. Explicable planning, introduced as a method to bridge this gap, aims to reconcile human expectations with the agent's optimal behavior, facilitating interpretable decision-making. However, an unresolved critical issue is ensuring safety in explicable planning, as it could result in explicable behaviors that are unsafe. To address this, we propose Safe Explicable Planning (SEP), which extends the prior work to support the specification of a safety bound. The goal of SEP is to find behaviors that align with human expectations while adhering to the specified safety criterion. Our approach generalizes the consideration of multiple objectives stemming from multiple models rather than a single model, yielding a Pareto set of safe explicable policies. We present both an exact method, guaranteeing finding the Pareto set, and a more efficient greedy method that finds one of the policies in the Pareto set. Additionally, we offer approximate solutions based on state aggregation to improve scalability. We provide formal proofs that validate the desired theoretical properties of these methods. Evaluation through simulations and physical robot experiments confirms the effectiveness of our approach for safe explicable planning.

♻ ☆ Resilient source seeking with robot swarms

Antonio Acuaviva, Jesus Bautista, Weijia Yao, Juan Jimenez, Hector Garcia de Marina

We present a solution for locating the source, or maximum, of an unknown scalar field using a swarm of mobile robots. Unlike relying on the traditional gradient information, the swarm determines an ascending direction to approach the source with arbitrary precision. The ascending direction is calculated from measurements of the field strength at the robot locations and their relative positions concerning the centroid. Rather than focusing on individual robots, we focus the analysis on the density of robots per unit area to guarantee a more resilient swarm, i.e., the functionality remains even if individuals go missing or are misplaced during the mission. We reinforce the robustness of the algorithm by providing sufficient conditions for the swarm shape so that the ascending direction is almost parallel to the gradient. The swarm can respond to an unexpected environment by morphing its shape and exploiting the existence of multiple ascending directions. Finally, we validate our approach numerically with hundreds of robots. The fact that a large number of robots always calculate an ascending direction compensates for the loss of individuals and mitigates issues arising from the actuator and sensor noises.

comment: 7 pages, submitted to CDC 2024

♻ ☆ Tuning-free Quasi-stiffness Control Framework of a Powered Transfemoral Prosthesis for Task-adaptive Walking RAL

Teng Ma, Shucong Yin, Zhimin Hou, Binxin Huang, Haoyong Yu, Chenglong Fu

Impedance-based control represents a prevalent strategy in the development of powered transfemoral prostheses. However, creating a task-adaptive, tuning-free controller that effectively generalizes across diverse locomotion modes and terrain conditions continues to be a significant challenge. This letter proposes a tuning-free and task-adaptive quasi-stiffness control framework for powered prostheses that generalizes across various walking tasks, including the torque-angle relationship reconstruction part and the quasi-stiffness controller design part. A Gaussian Process Regression (GPR) model is introduced to predict the target features of the human joint angle and torque in a new task. Subsequently, a Kernelized Movement Primitives (KMP) is employed to reconstruct the torque-angle relationship of the new task from multiple human reference trajectories and estimated target features. Based on the torque-angle relationship of the new task, a quasi-stiffness control approach is designed for a powered prosthesis. Finally, the proposed framework is validated through practical examples, including varying speeds and inclines walking tasks. Notably, the proposed framework not only aligns with but frequently surpasses the performance of a benchmark finite state machine impedance controller (FSMIC) without necessitating manual impedance tuning and has the potential to expand to variable walking tasks in daily life for the transfemoral amputees.

comment: 8 pages, 10 figures. This work has been submitted to the IEEE-RAL for possible publication

♻ ☆ Domain Randomization via Entropy Maximization ICLR 2024

Gabriele Tiboni, Pascal Klink, Jan Peters, Tatiana Tommasi, Carlo D'Eramo, Georgia Chalvatzaki

Varying dynamics parameters in simulation is a popular Domain Randomization (DR) approach for overcoming the reality gap in Reinforcement Learning (RL). Nevertheless, DR heavily hinges on the choice of the sampling distribution of the dynamics parameters, since high variability is crucial to regularize the agent's behavior but notoriously leads to overly conservative policies when randomizing excessively. In this paper, we propose a novel approach to address sim-to-real transfer, which automatically shapes dynamics distributions during training in simulation without requiring real-world data. We introduce DOmain RAndomization via Entropy MaximizatiON (DORAEMON), a constrained optimization problem that directly maximizes the entropy of the training distribution while retaining generalization capabilities. In achieving this, DORAEMON gradually increases the diversity of sampled dynamics parameters as long as the probability of success of the current policy is sufficiently high. We empirically validate the consistent benefits of DORAEMON in obtaining highly adaptive and generalizable policies, i.e. solving the task at hand across the widest range of dynamics parameters, as opposed to representative baselines from the DR literature. Notably, we also demonstrate the Sim2Real applicability of DORAEMON through its successful zero-shot transfer in a robotic manipulation setup under unknown real-world parameters.

comment: Published as a conference paper at ICLR 2024. Project website at https://gabrieletiboni.github.io/doraemon/

♻ ☆ SGS-SLAM: Semantic Gaussian Splatting For Neural Dense SLAM

Mingrui Li, Shuhong Liu, Heng Zhou, Guohao Zhu, Na Cheng, Tianchen Deng, Hongyu Wang

We present SGS-SLAM, the first semantic visual SLAM system based on Gaussian Splatting. It incorporates appearance, geometry, and semantic features through multi-channel optimization, addressing the oversmoothing limitations of neural implicit SLAM systems in high-quality rendering, scene understanding, and object-level geometry. We introduce a unique semantic feature loss that effectively compensates for the shortcomings of traditional depth and color losses in object optimization. Through a semantic-guided keyframe selection strategy, we prevent erroneous reconstructions caused by cumulative errors. Extensive experiments demonstrate that SGS-SLAM delivers state-of-the-art performance in camera pose estimation, map reconstruction, precise semantic segmentation, and object-level geometric accuracy, while ensuring real-time rendering capabilities.

♻ ☆ Towards Source-free Domain Adaptive Semantic Segmentation via Importance-aware and Prototype-contrast Learning

Yihong Cao, Hui Zhang, Xiao Lu, Zheng Xiao, Kailun Yang, Yaonan Wang

Domain adaptive semantic segmentation enables robust pixel-wise understanding in real-world driving scenes. Source-free domain adaptation, as a more practical technique, addresses the concerns of data privacy and storage limitations in typical unsupervised domain adaptation methods, making it especially relevant in the context of intelligent vehicles. It utilizes a well-trained source model and unlabeled target data to achieve adaptation in the target domain. However, in the absence of source data and target labels, current solutions cannot sufficiently reduce the impact of domain shift and fully leverage the information from the target data. In this paper, we propose an end-to-end source-free domain adaptation semantic segmentation method via Importance-Aware and Prototype-Contrast (IAPC) learning. The proposed IAPC framework effectively extracts domain-invariant knowledge from the well-trained source model and learns domain-specific knowledge from the unlabeled target domain. Specifically, considering the problem of domain shift in the prediction of the target domain by the source model, we put forward an importance-aware mechanism for the biased target prediction probability distribution to extract domain-invariant knowledge from the source model. We further introduce a prototype-contrast strategy, which includes a prototype-symmetric cross-entropy loss and a prototype-enhanced cross-entropy loss, to learn target intra-domain knowledge without relying on labels. A comprehensive variety of experiments on two domain adaptive semantic segmentation benchmarks demonstrates that the proposed end-to-end IAPC solution outperforms existing state-of-the-art methods. The source code is publicly available at https://github.com/yihong-97/Source-free-IAPC.

comment: Accepted to IEEE Transactions on Intelligent Vehicles (T-IV). The source code is publicly available at https://github.com/yihong-97/Source-free-IAPC

♻ ☆ When Robotics Meets Wireless Communications: An Introductory Tutorial

Daniel Bonilla Licea, Mounir Ghogho, Martin Saska

The importance of ground Mobile Robots (MRs) and Unmanned Aerial Vehicles (UAVs) within the research community, industry, and society is growing fast. Many of these agents are nowadays equipped with communication systems that are, in some cases, essential to successfully achieve certain tasks. In this context, we have begun to witness the development of a new interdisciplinary research field at the intersection of robotics and communications. This research field has been boosted by the intention of integrating UAVs within the 5G and 6G communication networks. This research will undoubtedly lead to many important applications in the near future. Nevertheless, one of the main obstacles to the development of this research area is that most researchers address these problems by oversimplifying either the robotics or the communications aspect. This impedes the ability of reaching the full potential of this new interdisciplinary research area. In this tutorial, we present some of the modelling tools necessary to address problems involving both robotics and communication from an interdisciplinary perspective. As an illustrative example of such problems, we focus in this tutorial on the issue of communication-aware trajectory planning.

comment: 35 pages, 192 references

♻ ☆ Motion Generation from Fine-grained Textual Descriptions

Kunhang Li, Yansong Feng

The task of text2motion is to generate human motion sequences from given textual descriptions, where the model explores diverse mappings from natural language instructions to human body movements. While most existing works are confined to coarse-grained motion descriptions, e.g., "A man squats.", fine-grained descriptions specifying movements of relevant body parts are barely explored. Models trained with coarse-grained texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure to generate motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset specializing in fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with step-by-step instructions with pseudo-code compulsory checks. Accordingly, we design a new text2motion model, FineMotionDiffuse, making full use of fine-grained textual information. Our quantitative evaluation shows that FineMotionDiffuse trained on FineHumanML3D improves FID by a large margin of 0.38, compared with competitive baselines. According to the qualitative evaluation and case study, our model outperforms MotionDiffuse in generating spatially or chronologically composite motions, by learning the implicit mappings from fine-grained descriptions to the corresponding basic motions. We release our data at https://github.com/KunhangL/finemotiondiffuse.

♻ ☆ Attention-based Estimation and Prediction of Human Intent to augment Haptic Glove aided Control of Robotic Hand

Muneeb Ahmed, Rajesh Kumar, Qaim Abbas, Brejesh Lall, Arzad A. Kherani, Sudipto Mukherjee

The letter focuses on Haptic Glove (HG) based control of a Robotic Hand (RH) executing in-hand manipulation of certain objects of interest. The high dimensional motion signals in HG and RH possess intrinsic variability of kinematics resulting in difficulty to establish a direct mapping of the motion signals from HG onto the RH. An estimation mechanism is proposed to quantify the motion signal acquired from the human controller in relation to the intended goal pose of the object being held by the robotic hand. A control algorithm is presented to transform the synthesized intent at the RH and allow relocation of the object to the expected goal pose. The lag in synthesis of the intent in the presence of communication delay leads to a requirement of predicting the estimated intent. We leverage an attention-based convolutional neural network encoder to predict the trajectory of intent for a certain lookahead to compensate for the delays. The proposed methodology is evaluated across objects of different shapes, mass, and materials. We present a comparative performance of the estimation and prediction mechanisms on 5G-driven real-world robotic setup against benchmark methodologies. The test-MSE in prediction of human intent is reported to yield ~ 97.3 -98.7% improvement of accuracy in comparison to LSTM-based benchmark

♻ ☆ Full Attitude Intelligent Controller Design of a Heliquad under Complete Failure of an Actuator

Eeshan Kulkarni, Suresh Sundaram

In this paper, we design a reliable Heliquad and develop an intelligent controller to handle one actuators complete failure. Heliquad is a multi-copter similar to Quadcopter, with four actuators diagonally symmetric from the center. Each actuator has two control inputs; the first input changes the propeller blades collective pitch (also called variable pitch), and the other input changes the rotation speed. For reliable operation and high torque characteristic requirement for yaw control, a cambered airfoil is used to design propeller blades. A neural network-based control allocation is designed to provide complete control authority even under a complete loss of one actuator. Nonlinear quaternion based outer loop position control, with proportional-derivative inner loop for attitude control and neural network-based control allocation is used in controller design. The proposed controller and Heliquad designs performance is evaluated using a software-in-loop simulation to track the position reference command under failure. The results clearly indicate that the Heliquad with an intelligent controller provides necessary tracking performance even under a complete loss of one actuator.

comment: 7 pages, For video go to https://indianinstituteofscience-my.sharepoint.com/:v:/g/personal/eeshank_iisc_ac_in/EcMg2uTtE91AsHDejNkb6YMBNckaXGjeh_YMzDV6sAHZAQ?e=DrRqmN

♻ ☆ Autonomous Hook-Based Grasping and Transportation with Quadcopters

Péter Antal, Tamás Péni, Roland Tóth

Payload grasping and transportation with quadcopters is an active research area that has rapidly developed over the last decade. To grasp a payload without human interaction, most state-of-the-art approaches apply robotic arms that are attached to the quadcopter body. However, due to the large weight and power consumption of these aerial manipulators, their agility and flight time are limited. This paper proposes a motion control and planning method for transportation with a lightweight, passive manipulator structure that consists of a hook attached to a quadrotor using a 1 DoF revolute joint. To perform payload grasping, transportation, and release, first, time-optimal reference trajectories are designed through specific waypoints to ensure the fast and reliable execution of the tasks. Then, a two-stage motion control approach is developed based on a robust geometric controller for precise and reliable reference tracking and a linear--quadratic payload regulator for rapid setpoint stabilization of the payload swing. Furthermore, stability of the closed-loop system is mathematically proven to give safety guarantee for its operation. The proposed control architecture and design are evaluated in a high-fidelity physical simulator, and also in real flight experiments, using a custom-made quadrotor--hook manipulator platform.

♻ ☆ Robustness Evaluation of Localization Techniques for Autonomous Racing

Tian Yi Lim, Edoardo Ghignone, Nicolas Baumann, Michele Magno

This work introduces SynPF, an MCL-based algorithm tailored for high-speed racing environments. Benchmarked against Cartographer, a state-of-the-art pose-graph SLAM algorithm, SynPF leverages synergies from previous particle-filtering methods and synthesizes them for the high-performance racing domain. Our extensive in-field evaluations reveal that while Cartographer excels under nominal conditions, it struggles when subjected to wheel-slip, a common phenomenon in a racing scenario due to varying grip levels and aggressive driving behaviour. Conversely, SynPF demonstrates robustness in these challenging conditions and a low-latency computation time of 1.25 ms on on-board computers without a GPU. Using the F1TENTH platform, a 1:10 scaled autonomous racing vehicle, this work not only highlights the vulnerabilities of existing algorithms in high-speed scenarios, tested up until 7.6 m/s, but also emphasizes the potential of SynPF as a viable alternative, especially in deteriorating odometry conditions.

comment: Accepted at the Design, Automation and Test in Europe Conference 2024 as an extended abstract

♻ ☆ OCC-VO: Dense Mapping via 3D Occupancy-Based Visual Odometry for Autonomous Driving

Heng Li, Yifan Duan, Xinran Zhang, Haiyi Liu, Jianmin Ji, Yanyong Zhang

Visual Odometry (VO) plays a pivotal role in autonomous systems, with a principal challenge being the lack of depth information in camera images. This paper introduces OCC-VO, a novel framework that capitalizes on recent advances in deep learning to transform 2D camera images into 3D semantic occupancy, thereby circumventing the traditional need for concurrent estimation of ego poses and landmark locations. Within this framework, we utilize the TPV-Former to convert surround view cameras' images into 3D semantic occupancy. Addressing the challenges presented by this transformation, we have specifically tailored a pose estimation and mapping algorithm that incorporates Semantic Label Filter, Dynamic Object Filter, and finally, utilizes Voxel PFilter for maintaining a consistent global semantic map. Evaluations on the Occ3D-nuScenes not only showcase a 20.6% improvement in Success Ratio and a 29.6% enhancement in trajectory accuracy against ORB-SLAM3, but also emphasize our ability to construct a comprehensive map. Our implementation is open-sourced and available at: https://github.com/USTCLH/OCC-VO.

comment: 7pages, 3 figures

♻ ☆ Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models

Joao Carvalho, An T. Le, Mark Baierl, Dorothea Koert, Jan Peters

Learning priors on trajectory distributions can help accelerate robot motion planning optimization. Given previously successful plans, learning trajectory generative models as priors for a new planning problem is highly desirable. Prior works propose several ways on utilizing this prior to bootstrapping the motion planning problem. Either sampling the prior for initializations or using the prior distribution in a maximum-a-posterior formulation for trajectory optimization. In this work, we propose learning diffusion models as priors. We then can sample directly from the posterior trajectory distribution conditioned on task goals, by leveraging the inverse denoising process of diffusion models. Furthermore, diffusion has been recently shown to effectively encode data multimodality in high-dimensional settings, which is particularly well-suited for large trajectory dataset. To demonstrate our method efficacy, we compare our proposed method - Motion Planning Diffusion - against several baselines in simulated planar robot and 7-dof robot arm manipulator environments. To assess the generalization capabilities of our method, we test it in environments with previously unseen obstacles. Our experiments show that diffusion models are strong priors to encode high-dimensional trajectory distributions of robot motions.

♻ ☆ Feeling Optimistic? Ambiguity Attitudes for Online Decision Making

Jared J. Beard, R. Michael Butts, Yu Gu

Due to the complexity of many decision making problems, tree search algorithms often have inadequate information to produce accurate transition models. Robust methods, designed to make safe decisions when faced with these uncertainties, often overlook the impact expressions of uncertainty have on how the decision is made. This work introduces the Ambiguity Attitude Graph Search (AAGS), advocating for more precise representation of ambiguities (uncertainty from a set of plausible models) in decision making. Additionally, AAGS allows users to adjust their ambiguity attitude (or preference), promoting exploration and improving users' ability to control how an agent should respond when faced with a set of valid alternatives. Simulation in a dynamic sailing environment shows how highly stochastic environments can lead robust methods to fail. Results further demonstrate how adjusting ambiguity attitudes better fulfills objectives while mitigating this failure mode of robust approaches. Because this approach is a generalization of the robust framework, these results further demonstrate how algorithms focused on ambiguity have applicability beyond safety-critical systems.

comment: 6 pages, 5 figures, 2 algorithms. Submitted to the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems in Abu Dhabi, UAE (Oct 14-18, 2024)

♻ ☆ Visual Whole-Body Control for Legged Loco-Manipulation

Minghuan Liu, Zixuan Chen, Xuxin Cheng, Yandong Ji, Ruihan Yang, Xiaolong Wang

We study the problem of mobile manipulation using legged robots equipped with an arm, namely legged loco-manipulation. The robot legs, while usually utilized for mobility, offer an opportunity to amplify the manipulation capabilities by conducting whole-body control. That is, the robot can control the legs and the arm at the same time to extend its workspace. We propose a framework that can conduct the whole-body control autonomously with visual observations. Our approach, namely Visual Whole-Body Control(VBC), is composed of a low-level policy using all degrees of freedom to track the end-effector manipulator position and a high-level policy proposing the end-effector position based on visual inputs. We train both levels of policies in simulation and perform Sim2Real transfer for real robot deployment. We perform extensive experiments and show significant improvements over baselines in picking up diverse objects in different configurations (heights, locations, orientations) and environments. Project page: https://wholebody-b1.github.io

comment: The first two authors contribute equally. Project page: https://wholebody-b1.github.io

☆ Efficient Video Object Segmentation via Modulated Cross-Attention Memory

Abdelrahman Shaker, Syed Talal Wasim, Martin Danelljan, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan

Recently, transformer-based approaches have shown promising results for semi-supervised video object segmentation. However, these approaches typically struggle on long videos due to increased GPU memory demands, as they frequently expand the memory bank every few frames. We propose a transformer-based approach, named MAVOS, that introduces an optimized and dynamic long-term modulated cross-attention (MCA) memory to model temporal smoothness without requiring frequent memory expansion. The proposed MCA effectively encodes both local and global features at various levels of granularity while efficiently maintaining consistent speed regardless of the video length. Extensive experiments on multiple benchmarks, LVOS, Long-Time Video, and DAVIS 2017, demonstrate the effectiveness of our proposed contributions leading to real-time inference and markedly reduced memory demands without any degradation in segmentation accuracy on long videos. Compared to the best existing transformer-based approach, our MAVOS increases the speed by 7.6x, while significantly reducing the GPU memory by 87% with comparable segmentation performance on short and long video datasets. Notably on the LVOS dataset, our MAVOS achieves a J&F score of 63.3% while operating at 37 frames per second (FPS) on a single V100 GPU. Our code and models will be publicly available at: https://github.com/Amshaker/MAVOS.

☆ ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis CVPR 2024

Muhammad Hamza Mughal, Rishabh Dabral, Ikhsanul Habibie, Lucia Donatelli, Marc Habermann, Christian Theobalt

Gestures play a key role in human communication. Recent methods for co-speech gesture generation, while managing to generate beat-aligned motions, struggle generating gestures that are semantically aligned with the utterance. Compared to beat gestures that align naturally to the audio signal, semantically coherent gestures require modeling the complex interactions between the language and human motion, and can be controlled by focusing on certain words. Therefore, we present ConvoFusion, a diffusion-based approach for multi-modal gesture synthesis, which can not only generate gestures based on multi-modal speech inputs, but can also facilitate controllability in gesture synthesis. Our method proposes two guidance objectives that allow the users to modulate the impact of different conditioning modalities (e.g. audio vs text) as well as to choose certain words to be emphasized during gesturing. Our method is versatile in that it can be trained either for generating monologue gestures or even the conversational gestures. To further advance the research on multi-party interactive gestures, the DnD Group Gesture dataset is released, which contains 6 hours of gesture data showing 5 people interacting with one another. We compare our method with several recent works and demonstrate effectiveness of our method on a variety of tasks. We urge the reader to watch our supplementary video at our website.

comment: CVPR 2024. Project Page: https://vcai.mpi-inf.mpg.de/projects/ConvoFusion/

☆ OmniVid: A Generative Framework for Universal Video Understanding CVPR 2024

Junke Wang, Dongdong Chen, Chong Luo, Bo He, Lu Yuan, Zuxuan Wu, Yu-Gang Jiang

The core of video understanding tasks, such as recognition, captioning, and tracking, is to automatically detect objects or actions in a video and analyze their temporal evolution. Despite sharing a common goal, different tasks often rely on distinct model architectures and annotation formats. In contrast, natural language processing benefits from a unified output space, i.e., text sequences, which simplifies the training of powerful foundational language models, such as GPT-3, with extensive training corpora. Inspired by this, we seek to unify the output space of video understanding tasks by using languages as labels and additionally introducing time and box tokens. In this way, a variety of video tasks could be formulated as video-grounded token generation. This enables us to address various types of video tasks, including classification (such as action recognition), captioning (covering clip captioning, video question answering, and dense video captioning), and localization tasks (such as visual object tracking) within a fully shared encoder-decoder architecture, following a generative framework. Through comprehensive experiments, we demonstrate such a simple and straightforward idea is quite effective and can achieve state-of-the-art or competitive results on seven video benchmarks, providing a novel perspective for more universal video understanding. Code is available at https://github.com/wangjk666/OmniVid.

comment: Accepted by CVPR 2024

☆ AiOS: All-in-One-Stage Expressive Human Pose and Shape Estimation

Qingping Sun, Yanjun Wang, Ailing Zeng, Wanqi Yin, Chen Wei, Wenjia Wang, Haiyi Mei, Chi Sing Leung, Ziwei Liu, Lei Yang, Zhongang Cai

Expressive human pose and shape estimation (a.k.a. 3D whole-body mesh recovery) involves the human body, hand, and expression estimation. Most existing methods have tackled this task in a two-stage manner, first detecting the human body part with an off-the-shelf detection model and inferring the different human body parts individually. Despite the impressive results achieved, these methods suffer from 1) loss of valuable contextual information via cropping, 2) introducing distractions, and 3) lacking inter-association among different persons and body parts, inevitably causing performance degradation, especially for crowded scenes. To address these issues, we introduce a novel all-in-one-stage framework, AiOS, for multiple expressive human pose and shape recovery without an additional human detection step. Specifically, our method is built upon DETR, which treats multi-person whole-body mesh recovery task as a progressive set prediction problem with various sequential detection. We devise the decoder tokens and extend them to our task. Specifically, we first employ a human token to probe a human location in the image and encode global features for each instance, which provides a coarse location for the later transformer block. Then, we introduce a joint-related token to probe the human joint in the image and encoder a fine-grained local feature, which collaborates with the global feature to regress the whole-body mesh. This straightforward but effective model outperforms previous state-of-the-art methods by a 9% reduction in NMVE on AGORA, a 30% reduction in PVE on EHF, a 10% reduction in PVE on ARCTIC, and a 3% reduction in PVE on EgoBody.

comment: Homepage: https://ttxskk.github.io/AiOS/

☆ SLEDGE: Synthesizing Simulation Environments for Driving Agents with Generative Models

Kashyap Chitta, Daniel Dauner, Andreas Geiger

SLEDGE is the first generative simulator for vehicle motion planning trained on real-world driving logs. Its core component is a learned model that is able to generate agent bounding boxes and lane graphs. The model's outputs serve as an initial state for traffic simulation. The unique properties of the entities to be generated for SLEDGE, such as their connectivity and variable count per scene, render the naive application of most modern generative models to this task non-trivial. Therefore, together with a systematic study of existing lane graph representations, we introduce a novel raster-to-vector autoencoder (RVAE). It encodes agents and the lane graph into distinct channels in a rasterized latent map. This facilitates both lane-conditioned agent generation and combined generation of lanes and agents with a Diffusion Transformer. Using generated entities in SLEDGE enables greater control over the simulation, e.g. upsampling turns or increasing traffic density. Further, SLEDGE can support 500m long routes, a capability not found in existing data-driven simulators like nuPlan. It presents new challenges for planning algorithms, evidenced by failure rates of over 40% for PDM, the winner of the 2023 nuPlan challenge, when tested on hard routes and dense traffic generated by our model. Compared to nuPlan, SLEDGE requires 500$\times$ less storage to set up (<4GB), making it a more accessible option and helping with democratizing future research in this field.

☆ Track Everything Everywhere Fast and Robustly

Yunzhou Song, Jiahui Lei, Ziyun Wang, Lingjie Liu, Kostas Daniilidis

We propose a novel test-time optimization approach for efficiently and robustly tracking any pixel at any time in a video. The latest state-of-the-art optimization-based tracking technique, OmniMotion, requires a prohibitively long optimization time, rendering it impractical for downstream applications. OmniMotion is sensitive to the choice of random seeds, leading to unstable convergence. To improve efficiency and robustness, we introduce a novel invertible deformation network, CaDeX++, which factorizes the function representation into a local spatial-temporal feature grid and enhances the expressivity of the coupling blocks with non-linear functions. While CaDeX++ incorporates a stronger geometric bias within its architectural design, it also takes advantage of the inductive bias provided by the vision foundation models. Our system utilizes monocular depth estimation to represent scene geometry and enhances the objective by incorporating DINOv2 long-term semantics to regulate the optimization process. Our experiments demonstrate a substantial improvement in training speed (more than \textbf{10 times} faster), robustness, and accuracy in tracking over the SoTA optimization-based method OmniMotion.

comment: project page: https://timsong412.github.io/FastOmniTrack/

☆ Towards Explaining Hypercomplex Neural Networks

Eleonora Lopez, Eleonora Grassucci, Debora Capriotti, Danilo Comminiello

Hypercomplex neural networks are gaining increasing interest in the deep learning community. The attention directed towards hypercomplex models originates from several aspects, spanning from purely theoretical and mathematical characteristics to the practical advantage of lightweight models over conventional networks, and their unique properties to capture both global and local relations. In particular, a branch of these architectures, parameterized hypercomplex neural networks (PHNNs), has also gained popularity due to their versatility across a multitude of application domains. Nonetheless, only few attempts have been made to explain or interpret their intricacies. In this paper, we propose inherently interpretable PHNNs and quaternion-like networks, thus without the need for any post-hoc method. To achieve this, we define a type of cosine-similarity transform within the parameterized hypercomplex domain. This PHB-cos transform induces weight alignment with relevant input features and allows to reduce the model into a single linear transform, rendering it directly interpretable. In this work, we start to draw insights into how this unique branch of neural models operates. We observe that hypercomplex networks exhibit a tendency to concentrate on the shape around the main object of interest, in addition to the shape of the object itself. We provide a thorough analysis, studying single neurons of different layers and comparing them against how real-valued networks learn. The code of the paper is available at https://github.com/ispamm/HxAI.

comment: The paper has been accepted at IEEE WCCI 2024

☆ FastCAR: Fast Classification And Regression Multi-Task Learning via Task Consolidation for Modelling a Continuous Property Variable of Object Classes

Anoop Kini, Andreas Jansche, Timo Bernthaler, Gerhard Schneider

FastCAR is a novel task consolidation approach in Multi-Task Learning (MTL) for a classification and a regression task, despite task heterogeneity with only subtle correlation. It addresses object classification and continuous property variable regression, a crucial use case in science and engineering. FastCAR involves a labeling transformation approach that can be used with a single-task regression network architecture. FastCAR outperforms traditional MTL model families, parametrized in the landscape of architecture and loss weighting schemes, when learning of both tasks are collectively considered (classification accuracy of 99.54%, regression mean absolute percentage error of 2.3%). The experiments performed used an Advanced Steel Property dataset contributed by us. The dataset comprises 4536 images of 224x224 pixels, annotated with object classes and hardness properties that take continuous values. With the labeling transformation and single-task regression network architecture, FastCAR achieves reduced latency and time efficiency.

☆ AID: Attention Interpolation of Text-to-Image Diffusion

Qiyuan He, Jinghao Wang, Ziwei Liu, Angela Yao

Conditional diffusion models can create unseen images in various settings, aiding image interpolation. Interpolation in latent spaces is well-studied, but interpolation with specific conditions like text or poses is less understood. Simple approaches, such as linear interpolation in the space of conditions, often result in images that lack consistency, smoothness, and fidelity. To that end, we introduce a novel training-free technique named Attention Interpolation via Diffusion (AID). Our key contributions include 1) proposing an inner/outer interpolated attention layer; 2) fusing the interpolated attention with self-attention to boost fidelity; and 3) applying beta distribution to selection to increase smoothness. We also present a variant, Prompt-guided Attention Interpolation via Diffusion (PAID), that considers interpolation as a condition-dependent generative process. This method enables the creation of new images with greater consistency, smoothness, and efficiency, and offers control over the exact path of interpolation. Our approach demonstrates effectiveness for conceptual and spatial interpolation. Code and demo are available at https://github.com/QY-H00/attention-interpolation-diffusion.

☆ TC4D: Trajectory-Conditioned Text-to-4D Generation

Sherwin Bahmani, Xian Liu, Yifan Wang, Ivan Skorokhodov, Victor Rong, Ziwei Liu, Xihui Liu, Jeong Joon Park, Sergey Tulyakov, Gordon Wetzstein, Andrea Tagliasacchi, David B. Lindell

Recent techniques for text-to-4D generation synthesize dynamic 3D scenes using supervision from pre-trained text-to-video models. However, existing representations for motion, such as deformation models or time-dependent neural representations, are limited in the amount of motion they can generate-they cannot synthesize motion extending far beyond the bounding box used for volume rendering. The lack of a more flexible motion model contributes to the gap in realism between 4D generation methods and recent, near-photorealistic video generation models. Here, we propose TC4D: trajectory-conditioned text-to-4D generation, which factors motion into global and local components. We represent the global motion of a scene's bounding box using rigid transformation along a trajectory parameterized by a spline. We learn local deformations that conform to the global trajectory using supervision from a text-to-video model. Our approach enables the synthesis of scenes animated along arbitrary trajectories, compositional scene generation, and significant improvements to the realism and amount of generated motion, which we evaluate qualitatively and through a user study. Video results can be viewed on our website: https://sherwinbahmani.github.io/tc4d.

comment: Project Page: https://sherwinbahmani.github.io/tc4d

☆ CMP: Cooperative Motion Prediction with Multi-Agent Communication

Zhuoyuan Wu, Yuping Wang, Hengbo Ma, Zhaowei Li, Hang Qiu, Jiachen Li

The confluence of the advancement of Autonomous Vehicles (AVs) and the maturity of Vehicle-to-Everything (V2X) communication has enabled the capability of cooperative connected and automated vehicles (CAVs). Building on top of cooperative perception, this paper explores the feasibility and effectiveness of cooperative motion prediction. Our method, CMP, takes LiDAR signals as input to enhance tracking and prediction capabilities. Unlike previous work that focuses separately on either cooperative perception or motion prediction, our framework, to the best of our knowledge, is the first to address the unified problem where CAVs share information in both perception and prediction modules. Incorporated into our design is the unique capability to tolerate realistic V2X bandwidth limitations and transmission delays, while dealing with bulky perception representations. We also propose a prediction aggregation module, which unifies the predictions obtained by different CAVs and generates the final prediction. Through extensive experiments and ablation studies, we demonstrate the effectiveness of our method in cooperative perception, tracking, and motion prediction tasks. In particular, CMP reduces the average prediction error by 17.2\% with fewer missing detections compared with the no cooperation setting. Our work marks a significant step forward in the cooperative capabilities of CAVs, showcasing enhanced performance in complex scenarios.

☆ Leveraging Near-Field Lighting for Monocular Depth Estimation from Endoscopy Videos

Akshay Paruchuri, Samuel Ehrenstein, Shuxian Wang, Inbar Fried, Stephen M. Pizer, Marc Niethammer, Roni Sengupta

Monocular depth estimation in endoscopy videos can enable assistive and robotic surgery to obtain better coverage of the organ and detection of various health issues. Despite promising progress on mainstream, natural image depth estimation, techniques perform poorly on endoscopy images due to a lack of strong geometric features and challenging illumination effects. In this paper, we utilize the photometric cues, i.e., the light emitted from an endoscope and reflected by the surface, to improve monocular depth estimation. We first create two novel loss functions with supervised and self-supervised variants that utilize a per-pixel shading representation. We then propose a novel depth refinement network (PPSNet) that leverages the same per-pixel shading representation. Finally, we introduce teacher-student transfer learning to produce better depth maps from both synthetic data with supervision and clinical data with self-supervision. We achieve state-of-the-art results on the C3VD dataset while estimating high-quality depth maps from clinical data. Our code, pre-trained models, and supplementary materials can be found on our project page: https://ppsnet.github.io/

comment: 26 pages, 7 tables, 7 figures

☆ ELGC-Net: Efficient Local-Global Context Aggregation for Remote Sensing Change Detection

Mubashir Noman, Mustansar Fiaz, Hisham Cholakkal, Salman Khan, Fahad Shahbaz Khan

Deep learning has shown remarkable success in remote sensing change detection (CD), aiming to identify semantic change regions between co-registered satellite image pairs acquired at distinct time stamps. However, existing convolutional neural network and transformer-based frameworks often struggle to accurately segment semantic change regions. Moreover, transformers-based methods with standard self-attention suffer from quadratic computational complexity with respect to the image resolution, making them less practical for CD tasks with limited training data. To address these issues, we propose an efficient change detection framework, ELGC-Net, which leverages rich contextual information to precisely estimate change regions while reducing the model size. Our ELGC-Net comprises a Siamese encoder, fusion modules, and a decoder. The focus of our design is the introduction of an Efficient Local-Global Context Aggregator module within the encoder, capturing enhanced global context and local spatial information through a novel pooled-transpose (PT) attention and depthwise convolution, respectively. The PT attention employs pooling operations for robust feature extraction and minimizes computational cost with transposed attention. Extensive experiments on three challenging CD datasets demonstrate that ELGC-Net outperforms existing methods. Compared to the recent transformer-based CD approach (ChangeFormer), ELGC-Net achieves a 1.4% gain in intersection over union metric on the LEVIR-CD dataset, while significantly reducing trainable parameters. Our proposed ELGC-Net sets a new state-of-the-art performance in remote sensing change detection benchmarks. Finally, we also introduce ELGC-Net-LW, a lighter variant with significantly reduced computational complexity, suitable for resource-constrained settings, while achieving comparable performance. Project url https://github.com/techmn/elgcnet.

comment: accepted at IEEE TGRS

☆ Serpent: Scalable and Efficient Image Restoration via Multi-scale Structured State Space Models

Mohammad Shahab Sepehri, Zalan Fabian, Mahdi Soltanolkotabi

The landscape of computational building blocks of efficient image restoration architectures is dominated by a combination of convolutional processing and various attention mechanisms. However, convolutional filters are inherently local and therefore struggle at modeling long-range dependencies in images. On the other hand, attention excels at capturing global interactions between arbitrary image regions, however at a quadratic cost in image dimension. In this work, we propose Serpent, an architecture that leverages recent advances in state space models (SSMs) in its core computational block. SSMs, originally introduced for sequence modeling, can maintain a global receptive field with a favorable linear scaling in input size. Our preliminary results demonstrate that Serpent can achieve reconstruction quality on par with state-of-the-art techniques, while requiring orders of magnitude less compute (up to $150$ fold reduction in FLOPS) and a factor of up to $5\times$ less GPU memory while maintaining a compact model size.

comment: 7 pages, 5 figures, preliminary workshop submission of a comprehensive work to be released soon

☆ Octree-GS: Towards Consistent Real-time Rendering with LOD-Structured 3D Gaussians

Kerui Ren, Lihan Jiang, Tao Lu, Mulin Yu, Linning Xu, Zhangkai Ni, Bo Dai

The recent 3D Gaussian splatting (3D-GS) has shown remarkable rendering fidelity and efficiency compared to NeRF-based neural scene representations. While demonstrating the potential for real-time rendering, 3D-GS encounters rendering bottlenecks in large scenes with complex details due to an excessive number of Gaussian primitives located within the viewing frustum. This limitation is particularly noticeable in zoom-out views and can lead to inconsistent rendering speeds in scenes with varying details. Moreover, it often struggles to capture the corresponding level of details at different scales with its heuristic density control operation. Inspired by the Level-of-Detail (LOD) techniques, we introduce Octree-GS, featuring an LOD-structured 3D Gaussian approach supporting level-of-detail decomposition for scene representation that contributes to the final rendering results. Our model dynamically selects the appropriate level from the set of multi-resolution anchor points, ensuring consistent rendering performance with adaptive LOD adjustments while maintaining high-fidelity rendering results.

comment: Project page: https://city-super.github.io/octree-gs/

☆ A Survey on 3D Egocentric Human Pose Estimation

Md Mushfiqur Azam, Kevin Desai

Egocentric human pose estimation aims to estimate human body poses and develop body representations from a first-person camera perspective. It has gained vast popularity in recent years because of its wide range of applications in sectors like XR-technologies, human-computer interaction, and fitness tracking. However, to the best of our knowledge, there is no systematic literature review based on the proposed solutions regarding egocentric 3D human pose estimation. To that end, the aim of this survey paper is to provide an extensive overview of the current state of egocentric pose estimation research. In this paper, we categorize and discuss the popular datasets and the different pose estimation models, highlighting the strengths and weaknesses of different methods by comparative analysis. This survey can be a valuable resource for both researchers and practitioners in the field, offering insights into key concepts and cutting-edge solutions in egocentric pose estimation, its wide-ranging applications, as well as the open problems with future scope.

☆ 2D Gaussian Splatting for Geometrically Accurate Radiance Fields

Binbin Huang, Zehao Yu, Anpei Chen, Andreas Geiger, Shenghua Gao

3D Gaussian Splatting (3DGS) has recently revolutionized radiance field reconstruction, achieving high quality novel view synthesis and fast rendering speed without baking. However, 3DGS fails to accurately represent surfaces due to the multi-view inconsistent nature of 3D Gaussians. We present 2D Gaussian Splatting (2DGS), a novel approach to model and reconstruct geometrically accurate radiance fields from multi-view images. Our key idea is to collapse the 3D volume into a set of 2D oriented planar Gaussian disks. Unlike 3D Gaussians, 2D Gaussians provide view-consistent geometry while modeling surfaces intrinsically. To accurately recover thin surfaces and achieve stable optimization, we introduce a perspective-accurate 2D splatting process utilizing ray-splat intersection and rasterization. Additionally, we incorporate depth distortion and normal consistency terms to further enhance the quality of the reconstructions. We demonstrate that our differentiable renderer allows for noise-free and detailed geometry reconstruction while maintaining competitive appearance quality, fast training speed, and real-time rendering. Our code will be made publicly available.

comment: 12 pages, 12 figures

☆ Sen2Fire: A Challenging Benchmark Dataset for Wildfire Detection using Sentinel Data

Yonghao Xu, Amanda Berg, Leif Haglund

Utilizing satellite imagery for wildfire detection presents substantial potential for practical applications. To advance the development of machine learning algorithms in this domain, our study introduces the \textit{Sen2Fire} dataset--a challenging satellite remote sensing dataset tailored for wildfire detection. This dataset is curated from Sentinel-2 multi-spectral data and Sentinel-5P aerosol product, comprising a total of 2466 image patches. Each patch has a size of 512$\times$512 pixels with 13 bands. Given the distinctive sensitivities of various wavebands to wildfire responses, our research focuses on optimizing wildfire detection by evaluating different wavebands and employing a combination of spectral indices, such as normalized burn ratio (NBR) and normalized difference vegetation index (NDVI). The results suggest that, in contrast to using all bands for wildfire detection, selecting specific band combinations yields superior performance. Additionally, our study underscores the positive impact of integrating Sentinel-5 aerosol data for wildfire detection. The code and dataset are available online (https://zenodo.org/records/10881058).

☆ Superior and Pragmatic Talking Face Generation with Teacher-Student Framework

Chao Liang, Jianwen Jiang, Tianyun Zhong, Gaojie Lin, Zhengkun Rong, Jiaqi Yang, Yongming Zhu

Talking face generation technology creates talking videos from arbitrary appearance and motion signal, with the "arbitrary" offering ease of use but also introducing challenges in practical applications. Existing methods work well with standard inputs but suffer serious performance degradation with intricate real-world ones. Moreover, efficiency is also an important concern in deployment. To comprehensively address these issues, we introduce SuperFace, a teacher-student framework that balances quality, robustness, cost and editability. We first propose a simple but effective teacher model capable of handling inputs of varying qualities to generate high-quality results. Building on this, we devise an efficient distillation strategy to acquire an identity-specific student model that maintains quality with significantly reduced computational load. Our experiments validate that SuperFace offers a more comprehensive solution than existing methods for the four mentioned objectives, especially in reducing FLOPs by 99\% with the student model. SuperFace can be driven by both video and audio and allows for localized facial attributes editing.

☆ Deepfake Generation and Detection: A Benchmark and Survey

Gan Pei, Jiangning Zhang, Menghan Hu, Guangtao Zhai, Chengjie Wang, Zhenyu Zhang, Jian Yang, Chunhua Shen, Dacheng Tao

In addition to the advancements in deepfake generation, corresponding detection technologies need to continuously evolve to regulate the potential misuse of deepfakes, such as for privacy invasion and phishing attacks. This survey comprehensively reviews the latest developments in deepfake generation and detection, summarizing and analyzing the current state of the art in this rapidly evolving field. We first unify task definitions, comprehensively introduce datasets and metrics, and discuss the development of generation and detection technology frameworks. Then, we discuss the development of several related sub-fields and focus on researching four mainstream deepfake fields: popular face swap, face reenactment, talking face generation, and facial attribute editing, as well as foreign detection. Subsequently, we comprehensively benchmark representative methods on popular datasets for each field, fully evaluating the latest and influential works published in top conferences/journals. Finally, we analyze the challenges and future research directions of the discussed fields. We closely follow the latest developments in https://github.com/flyingby/Awesome-Deepfake-Generation-and-Detection.

☆ Low-Latency Neural Stereo Streaming CVPR2024

Qiqi Hou, Farzad Farhadzadeh, Amir Said, Guillaume Sautiere, Hoang Le

The rise of new video modalities like virtual reality or autonomous driving has increased the demand for efficient multi-view video compression methods, both in terms of rate-distortion (R-D) performance and in terms of delay and runtime. While most recent stereo video compression approaches have shown promising performance, they compress left and right views sequentially, leading to poor parallelization and runtime performance. This work presents Low-Latency neural codec for Stereo video Streaming (LLSS), a novel parallel stereo video coding method designed for fast and efficient low-latency stereo video streaming. Instead of using a sequential cross-view motion compensation like existing methods, LLSS introduces a bidirectional feature shifting module to directly exploit mutual information among views and encode them effectively with a joint cross-view prior model for entropy coding. Thanks to this design, LLSS processes left and right views in parallel, minimizing latency; all while substantially improving R-D performance compared to both existing neural and conventional codecs.

comment: Accepted by CVPR2024

☆ Boosting Diffusion Models with Moving Average Sampling in Frequency Domain CVPR 2024

Yurui Qian, Qi Cai, Yingwei Pan, Yehao Li, Ting Yao, Qibin Sun, Tao Mei

Diffusion models have recently brought a powerful revolution in image generation. Despite showing impressive generative capabilities, most of these models rely on the current sample to denoise the next one, possibly resulting in denoising instability. In this paper, we reinterpret the iterative denoising process as model optimization and leverage a moving average mechanism to ensemble all the prior samples. Instead of simply applying moving average to the denoised samples at different timesteps, we first map the denoised samples to data space and then perform moving average to avoid distribution shift across timesteps. In view that diffusion models evolve the recovery from low-frequency components to high-frequency details, we further decompose the samples into different frequency components and execute moving average separately on each component. We name the complete approach "Moving Average Sampling in Frequency domain (MASF)". MASF could be seamlessly integrated into mainstream pre-trained diffusion models and sampling schedules. Extensive experiments on both unconditional and conditional diffusion models demonstrate that our MASF leads to superior performances compared to the baselines, with almost negligible additional complexity cost.

comment: CVPR 2024

☆ To Supervise or Not to Supervise: Understanding and Addressing the Key Challenges of 3D Transfer Learning

Souhail Hadgi, Lei Li, Maks Ovsjanikov

Transfer learning has long been a key factor in the advancement of many fields including 2D image analysis. Unfortunately, its applicability in 3D data processing has been relatively limited. While several approaches for 3D transfer learning have been proposed in recent literature, with contrastive learning gaining particular prominence, most existing methods in this domain have only been studied and evaluated in limited scenarios. Most importantly, there is currently a lack of principled understanding of both when and why 3D transfer learning methods are applicable. Remarkably, even the applicability of standard supervised pre-training is poorly understood. In this work, we conduct the first in-depth quantitative and qualitative investigation of supervised and contrastive pre-training strategies and their utility in downstream 3D tasks. We demonstrate that layer-wise analysis of learned features provides significant insight into the downstream utility of trained networks. Informed by this analysis, we propose a simple geometric regularization strategy, which improves the transferability of supervised pre-training. Our work thus sheds light onto both the specific challenges of 3D transfer learning, as well as strategies to overcome them.

★ Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation

Abdelrhman Werby, Chenguang Huang, Martin Büchner, Abhinav Valada, Wolfram Burgard

Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features. While these maps allow for the prediction of point-wise saliency maps when queried for a certain language concept, large-scale environments and abstract queries beyond the object level still pose a considerable hurdle, ultimately limiting language-grounded robotic navigation. In this work, we present HOV-SG, a hierarchical open-vocabulary 3D scene graph mapping approach for language-grounded robot navigation. Leveraging open-vocabulary vision foundation models, we first obtain state-of-the-art open-vocabulary segment-level maps in 3D and subsequently construct a 3D scene graph hierarchy consisting of floor, room, and object concepts, each enriched with open-vocabulary features. Our approach is able to represent multi-story buildings and allows robotic traversal of those using a cross-floor Voronoi graph. HOV-SG is evaluated on three distinct datasets and surpasses previous baselines in open-vocabulary semantic accuracy on the object, room, and floor level while producing a 75% reduction in representation size compared to dense open-vocabulary maps. In order to prove the efficacy and generalization capabilities of HOV-SG, we showcase successful long-horizon language-conditioned robot navigation within real-world multi-storage environments. We provide code and trial video data at http://hovsg.github.io/.

comment: Code and video are available at http://hovsg.github.io/

☆ ReMamber: Referring Image Segmentation with Mamba Twister

Yuhuan Yang, Chaofan Ma, Jiangchao Yao, Zhun Zhong, Ya Zhang, Yanfeng Wang

Referring Image Segmentation (RIS) leveraging transformers has achieved great success on the interpretation of complex visual-language tasks. However, the quadratic computation cost makes it resource-consuming in capturing long-range visual-language dependencies. Fortunately, Mamba addresses this with efficient linear complexity in processing. However, directly applying Mamba to multi-modal interactions presents challenges, primarily due to inadequate channel interactions for the effective fusion of multi-modal data. In this paper, we propose ReMamber, a novel RIS architecture that integrates the power of Mamba with a multi-modal Mamba Twister block. The Mamba Twister explicitly models image-text interaction, and fuses textual and visual features through its unique channel and spatial twisting mechanism. We achieve the state-of-the-art on three challenging benchmarks. Moreover, we conduct thorough analyses of ReMamber and discuss other fusion designs using Mamba. These provide valuable perspectives for future research.

☆ GTA-HDR: A Large-Scale Synthetic Dataset for HDR Image Reconstruction

Hrishav Bakul Barua, Kalin Stefanov, KokSheik Wong, Abhinav Dhall, Ganesh Krishnasamy

High Dynamic Range (HDR) content (i.e., images and videos) has a broad range of applications. However, capturing HDR content from real-world scenes is expensive and time- consuming. Therefore, the challenging task of reconstructing visually accurate HDR images from their Low Dynamic Range (LDR) counterparts is gaining attention in the vision research community. A major challenge in this research problem is the lack of datasets, which capture diverse scene conditions (e.g., lighting, shadows, weather, locations, landscapes, objects, humans, buildings) and various image features (e.g., color, contrast, saturation, hue, luminance, brightness, radiance). To address this gap, in this paper, we introduce GTA-HDR, a large-scale synthetic dataset of photo-realistic HDR images sampled from the GTA-V video game. We perform thorough evaluation of the proposed dataset, which demonstrates significant qualitative and quantitative improvements of the state-of-the-art HDR image reconstruction methods. Furthermore, we demonstrate the effectiveness of the proposed dataset and its impact on additional computer vision tasks including 3D human pose estimation, human body part segmentation, and holistic scene segmentation. The dataset, data collection pipeline, and evaluation code are available at: https://github.com/HrishavBakulBarua/GTA-HDR.

comment: Submitted to IEEE

☆ A foundation model utilizing chest CT volumes and radiology reports for supervised-level zero-shot detection of abnormalities

Ibrahim Ethem Hamamci, Sezgin Er, Furkan Almas, Ayse Gulnihan Simsek, Sevval Nil Esirgun, Irem Dogan, Muhammed Furkan Dasdelen, Bastian Wittmann, Enis Simsar, Mehmet Simsar, Emine Bensu Erdemir, Abdullah Alanbay, Anjany Sekuboyina, Berkan Lafci, Mehmet K. Ozdemir, Bjoern Menze

A major challenge in computational research in 3D medical imaging is the lack of comprehensive datasets. Addressing this issue, our study introduces CT-RATE, the first 3D medical imaging dataset that pairs images with textual reports. CT-RATE consists of 25,692 non-contrast chest CT volumes, expanded to 50,188 through various reconstructions, from 21,304 unique patients, along with corresponding radiology text reports. Leveraging CT-RATE, we developed CT-CLIP, a CT-focused contrastive language-image pre-training framework. As a versatile, self-supervised model, CT-CLIP is designed for broad application and does not require task-specific training. Remarkably, CT-CLIP outperforms state-of-the-art, fully supervised methods in multi-abnormality detection across all key metrics, thus eliminating the need for manual annotation. We also demonstrate its utility in case retrieval, whether using imagery or textual queries, thereby advancing knowledge dissemination. The open-source release of CT-RATE and CT-CLIP marks a significant advancement in medical AI, enhancing 3D imaging analysis and fostering innovation in healthcare.

☆ Assessment of Multimodal Large Language Models in Alignment with Human Values

Zhelun Shi, Zhipin Wang, Hongxing Fan, Zaibin Zhang, Lijun Li, Yongting Zhang, Zhenfei Yin, Lu Sheng, Yu Qiao, Jing Shao

Large Language Models (LLMs) aim to serve as versatile assistants aligned with human values, as defined by the principles of being helpful, honest, and harmless (hhh). However, in terms of Multimodal Large Language Models (MLLMs), despite their commendable performance in perception and reasoning tasks, their alignment with human values remains largely unexplored, given the complexity of defining hhh dimensions in the visual world and the difficulty in collecting relevant data that accurately mirrors real-world situations. To address this gap, we introduce Ch3Ef, a Compreh3ensive Evaluation dataset and strategy for assessing alignment with human expectations. Ch3Ef dataset contains 1002 human-annotated data samples, covering 12 domains and 46 tasks based on the hhh principle. We also present a unified evaluation strategy supporting assessment across various scenarios and different perspectives. Based on the evaluation results, we summarize over 10 key findings that deepen the understanding of MLLM capabilities, limitations, and the dynamic relationships between evaluation levels, guiding future advancements in the field.

comment: arXiv admin note: text overlap with arXiv:2311.02692

☆ DiffH2O: Diffusion-Based Synthesis of Hand-Object Interactions from Textual Descriptions

Sammy Christen, Shreyas Hampali, Fadime Sener, Edoardo Remelli, Tomas Hodan, Eric Sauser, Shugao Ma, Bugra Tekin

Generating natural hand-object interactions in 3D is challenging as the resulting hand and object motions are expected to be physically plausible and semantically meaningful. Furthermore, generalization to unseen objects is hindered by the limited scale of available hand-object interaction datasets. We propose DiffH2O, a novel method to synthesize realistic, one or two-handed object interactions from provided text prompts and geometry of the object. The method introduces three techniques that enable effective learning from limited data. First, we decompose the task into a grasping stage and a text-based interaction stage and use separate diffusion models for each. In the grasping stage, the model only generates hand motions, whereas in the interaction phase both hand and object poses are synthesized. Second, we propose a compact representation that tightly couples hand and object poses. Third, we propose two different guidance schemes to allow more control of the generated motions: grasp guidance and detailed textual guidance. Grasp guidance takes a single target grasping pose and guides the diffusion model to reach this grasp at the end of the grasping stage, which provides control over the grasping pose. Given a grasping motion from this stage, multiple different actions can be prompted in the interaction phase. For textual guidance, we contribute comprehensive text descriptions to the GRAB dataset and show that they enable our method to have more fine-grained control over hand-object interactions. Our quantitative and qualitative evaluation demonstrates that the proposed method outperforms baseline methods and leads to natural hand-object motions. Moreover, we demonstrate the practicality of our framework by utilizing a hand pose estimate from an off-the-shelf pose estimator for guidance, and then sampling multiple different actions in the interaction stage.

comment: Project Page: https://diffh2o.github.io/

☆ Efficient Image Pre-Training with Siamese Cropped Masked Autoencoders

Alexandre Eymaël, Renaud Vandeghen, Anthony Cioppa, Silvio Giancola, Bernard Ghanem, Marc Van Droogenbroeck

Self-supervised pre-training of image encoders is omnipresent in the literature, particularly following the introduction of Masked autoencoders (MAE). Current efforts attempt to learn object-centric representations from motion in videos. In particular, SiamMAE recently introduced a Siamese network, training a shared-weight encoder from two frames of a video with a high asymmetric masking ratio (95%). In this work, we propose CropMAE, an alternative approach to the Siamese pre-training introduced by SiamMAE. Our method specifically differs by exclusively considering pairs of cropped images sourced from the same image but cropped differently, deviating from the conventional pairs of frames extracted from a video. CropMAE therefore alleviates the need for video datasets, while maintaining competitive performances and drastically reducing pre-training time. Furthermore, we demonstrate that CropMAE learns similar object-centric representations without explicit motion, showing that current self-supervised learning methods do not learn objects from motion, but rather thanks to the Siamese architecture. Finally, CropMAE achieves the highest masking ratio to date (98.5%), enabling the reconstruction of images using only two visible patches. Our code is available at https://github.com/alexandre-eymael/CropMAE.

comment: 19 pages, 6 figures, 3 tables, 1 page of supplementary material

☆ DN-Splatter: Depth and Normal Priors for Gaussian Splatting and Meshing

Matias Turkulainen, Xuqian Ren, Iaroslav Melekhov, Otto Seiskari, Esa Rahtu, Juho Kannala

3D Gaussian splatting, a novel differentiable rendering technique, has achieved state-of-the-art novel view synthesis results with high rendering speeds and relatively low training times. However, its performance on scenes commonly seen in indoor datasets is poor due to the lack of geometric constraints during optimization. We extend 3D Gaussian splatting with depth and normal cues to tackle challenging indoor datasets and showcase techniques for efficient mesh extraction, an important downstream application. Specifically, we regularize the optimization procedure with depth information, enforce local smoothness of nearby Gaussians, and use the geometry of the 3D Gaussians supervised by normal cues to achieve better alignment with the true scene geometry. We improve depth estimation and novel view synthesis results over baselines and show how this simple yet effective regularization technique can be used to directly extract meshes from the Gaussian representation yielding more physically accurate reconstructions on indoor scenes. Our code will be released in https://github.com/maturk/dn-splatter.

☆ Annotated Biomedical Video Generation using Denoising Diffusion Probabilistic Models and Flow Fields

Rüveyda Yilmaz, Dennis Eschweiler, Johannes Stegmaier

The segmentation and tracking of living cells play a vital role within the biomedical domain, particularly in cancer research, drug development, and developmental biology. These are usually tedious and time-consuming tasks that are traditionally done by biomedical experts. Recently, to automatize these processes, deep learning based segmentation and tracking methods have been proposed. These methods require large-scale datasets and their full potential is constrained by the scarcity of annotated data in the biomedical imaging domain. To address this limitation, we propose Biomedical Video Diffusion Model (BVDM), capable of generating realistic-looking synthetic microscopy videos. Trained only on a single real video, BVDM can generate videos of arbitrary length with pixel-level annotations that can be used for training data-hungry models. It is composed of a denoising diffusion probabilistic model (DDPM) generating high-fidelity synthetic cell microscopy images and a flow prediction model (FPM) predicting the non-rigid transformation between consecutive video frames. During inference, initially, the DDPM imposes realistic cell textures on synthetic cell masks which are generated based on real data statistics. The flow prediction model predicts the flow field between consecutive masks and applies that to the DDPM output from the previous time frame to create the next one while keeping temporal consistency. BVDM outperforms state-of-the-art synthetic live cell microscopy video generation models. Furthermore, we demonstrate that a sufficiently large synthetic dataset enhances the performance of cell segmentation and tracking models compared to using a limited amount of available real data.

☆ Improving Text-to-Image Consistency via Automatic Prompt Optimization

Oscar Mañas, Pietro Astolfi, Melissa Hall, Candace Ross, Jack Urbanek, Adina Williams, Aishwarya Agrawal, Adriana Romero-Soriano, Michal Drozdzal

Impressive advances in text-to-image (T2I) generative models have yielded a plethora of high performing models which are able to generate aesthetically appealing, photorealistic images. Despite the progress, these models still struggle to produce images that are consistent with the input prompt, oftentimes failing to capture object quantities, relations and attributes properly. Existing solutions to improve prompt-image consistency suffer from the following challenges: (1) they oftentimes require model fine-tuning, (2) they only focus on nearby prompt samples, and (3) they are affected by unfavorable trade-offs among image quality, representation diversity, and prompt-image consistency. In this paper, we address these challenges and introduce a T2I optimization-by-prompting framework, OPT2I, which leverages a large language model (LLM) to improve prompt-image consistency in T2I models. Our framework starts from a user prompt and iteratively generates revised prompts with the goal of maximizing a consistency score. Our extensive validation on two datasets, MSCOCO and PartiPrompts, shows that OPT2I can boost the initial consistency score by up to 24.9% in terms of DSG score while preserving the FID and increasing the recall between generated and real data. Our work paves the way toward building more reliable and robust T2I systems by harnessing the power of LLMs.

☆ Towards 3D Vision with Low-Cost Single-Photon Cameras

Fangzhou Mu, Carter Sifferman, Sacha Jungerman, Yiquan Li, Mark Han, Michael Gleicher, Mohit Gupta, Yin Li

We present a method for reconstructing 3D shape of arbitrary Lambertian objects based on measurements by miniature, energy-efficient, low-cost single-photon cameras. These cameras, operating as time resolved image sensors, illuminate the scene with a very fast pulse of diffuse light and record the shape of that pulse as it returns back from the scene at a high temporal resolution. We propose to model this image formation process, account for its non-idealities, and adapt neural rendering to reconstruct 3D geometry from a set of spatially distributed sensors with known poses. We show that our approach can successfully recover complex 3D shapes from simulated data. We further demonstrate 3D object reconstruction from real-world captures, utilizing measurements from a commodity proximity sensor. Our work draws a connection between image-based modeling and active range scanning and is a step towards 3D vision with single-photon cameras.

☆ Evaluating the Efficacy of Prompt-Engineered Large Multimodal Models Versus Fine-Tuned Vision Transformers in Image-Based Security Applications

Fouad Trad, Ali Chehab

The success of Large Language Models (LLMs) has led to a parallel rise in the development of Large Multimodal Models (LMMs), such as Gemini-pro, which have begun to transform a variety of applications. These sophisticated multimodal models are designed to interpret and analyze complex data, integrating both textual and visual information on a scale previously unattainable, opening new avenues for a range of applications. This paper investigates the applicability and effectiveness of prompt-engineered Gemini-pro LMMs versus fine-tuned Vision Transformer (ViT) models in addressing critical security challenges. We focus on two distinct tasks: a visually evident task of detecting simple triggers, such as small squares in images, indicative of potential backdoors, and a non-visually evident task of malware classification through visual representations. Our results highlight a significant divergence in performance, with Gemini-pro falling short in accuracy and reliability when compared to fine-tuned ViT models. The ViT models, on the other hand, demonstrate exceptional accuracy, achieving near-perfect performance on both tasks. This study not only showcases the strengths and limitations of prompt-engineered LMMs in cybersecurity applications but also emphasizes the unmatched efficacy of fine-tuned ViT models for precise and dependable tasks.

☆ GenesisTex: Adapting Image Denoising Diffusion to Texture Space

Chenjian Gao, Boyan Jiang, Xinghui Li, Yingpeng Zhang, Qian Yu

We present GenesisTex, a novel method for synthesizing textures for 3D geometries from text descriptions. GenesisTex adapts the pretrained image diffusion model to texture space by texture space sampling. Specifically, we maintain a latent texture map for each viewpoint, which is updated with predicted noise on the rendering of the corresponding viewpoint. The sampled latent texture maps are then decoded into a final texture map. During the sampling process, we focus on both global and local consistency across multiple viewpoints: global consistency is achieved through the integration of style consistency mechanisms within the noise prediction network, and low-level consistency is achieved by dynamically aligning latent textures. Finally, we apply reference-based inpainting and img2img on denser views for texture refinement. Our approach overcomes the limitations of slow optimization in distillation-based methods and instability in inpainting-based methods. Experiments on meshes from various sources demonstrate that our method surpasses the baseline methods quantitatively and qualitatively.

comment: 12 pages, 10 figures

☆ CT Synthesis with Conditional Diffusion Models for Abdominal Lymph Node Segmentation

Yongrui Yu, Hanyu Chen, Zitian Zhang, Qiong Xiao, Wenhui Lei, Linrui Dai, Yu Fu, Hui Tan, Guan Wang, Peng Gao, Xiaofan Zhang

Despite the significant success achieved by deep learning methods in medical image segmentation, researchers still struggle in the computer-aided diagnosis of abdominal lymph nodes due to the complex abdominal environment, small and indistinguishable lesions, and limited annotated data. To address these problems, we present a pipeline that integrates the conditional diffusion model for lymph node generation and the nnU-Net model for lymph node segmentation to improve the segmentation performance of abdominal lymph nodes through synthesizing a diversity of realistic abdominal lymph node data. We propose LN-DDPM, a conditional denoising diffusion probabilistic model (DDPM) for lymph node (LN) generation. LN-DDPM utilizes lymph node masks and anatomical structure masks as model conditions. These conditions work in two conditioning mechanisms: global structure conditioning and local detail conditioning, to distinguish between lymph nodes and their surroundings and better capture lymph node characteristics. The obtained paired abdominal lymph node images and masks are used for the downstream segmentation task. Experimental results on the abdominal lymph node datasets demonstrate that LN-DDPM outperforms other generative methods in the abdominal lymph node image synthesis and better assists the downstream abdominal lymph node segmentation task.

☆ MUTE-SLAM: Real-Time Neural SLAM with Multiple Tri-Plane Hash Representations

Yifan Yan, Ruomin He, Zhenghua Liu

We introduce MUTE-SLAM, a real-time neural RGB-D SLAM system employing multiple tri-plane hash-encodings for efficient scene representation. MUTE-SLAM effectively tracks camera positions and incrementally builds a scalable multi-map representation for both small and large indoor environments. It dynamically allocates sub-maps for newly observed local regions, enabling constraint-free mapping without prior scene information. Unlike traditional grid-based methods, we use three orthogonal axis-aligned planes for hash-encoding scene properties, significantly reducing hash collisions and the number of trainable parameters. This hybrid approach not only speeds up convergence but also enhances the fidelity of surface reconstruction. Furthermore, our optimization strategy concurrently optimizes all sub-maps intersecting with the current camera frustum, ensuring global consistency. Extensive testing on both real-world and synthetic datasets has shown that MUTE-SLAM delivers state-of-the-art surface reconstruction quality and competitive tracking performance across diverse indoor settings. The code will be made public upon acceptance of the paper.

☆ Makeup Prior Models for 3D Facial Makeup Estimation and Applications CVPR2024

Xingchao Yang, Takafumi Taketomi, Yuki Endo, Yoshihiro Kanamori

In this work, we introduce two types of makeup prior models to extend existing 3D face prior models: PCA-based and StyleGAN2-based priors. The PCA-based prior model is a linear model that is easy to construct and is computationally efficient. However, it retains only low-frequency information. Conversely, the StyleGAN2-based model can represent high-frequency information with relatively higher computational cost than the PCA-based model. Although there is a trade-off between the two models, both are applicable to 3D facial makeup estimation and related applications. By leveraging makeup prior models and designing a makeup consistency module, we effectively address the challenges that previous methods faced in robustly estimating makeup, particularly in the context of handling self-occluded faces. In experiments, we demonstrate that our approach reduces computational costs by several orders of magnitude, achieving speeds up to 180 times faster. In addition, by improving the accuracy of the estimated makeup, we confirm that our methods are highly advantageous for various 3D facial makeup applications such as 3D makeup face reconstruction, user-friendly makeup editing, makeup transfer, and interpolation.

comment: CVPR2024. Project: https://yangxingchao.github.io/makeup-priors-page

☆ Noise2Noise Denoising of CRISM Hyperspectral Data ICLR 2024

Robert Platt, Rossella Arcucci, Cédric M. John

Hyperspectral data acquired by the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) have allowed for unparalleled mapping of the surface mineralogy of Mars. Due to sensor degradation over time, a significant portion of the recently acquired data is considered unusable. Here a new data-driven model architecture, Noise2Noise4Mars (N2N4M), is introduced to remove noise from CRISM images. Our model is self-supervised and does not require zero-noise target data, making it well suited for use in Planetary Science applications where high quality labelled data is scarce. We demonstrate its strong performance on synthetic-noise data and CRISM images, and its impact on downstream classification performance, outperforming benchmark methods on most metrics. This allows for detailed analysis for critical sites of interest on the Martian surface, including proposed lander sites.

comment: 5 pages, 3 figures. Accepted as a conference paper at the ICLR 2024 ML4RS Workshop

★ DataCook: Crafting Anti-Adversarial Examples for Healthcare Data Copyright Protection

Sihan Shang, Jiancheng Yang, Zhenglong Sun, Pascal Fua

In the realm of healthcare, the challenges of copyright protection and unauthorized third-party misuse are increasingly significant. Traditional methods for data copyright protection are applied prior to data distribution, implying that models trained on these data become uncontrollable. This paper introduces a novel approach, named DataCook, designed to safeguard the copyright of healthcare data during the deployment phase. DataCook operates by "cooking" the raw data before distribution, enabling the development of models that perform normally on this processed data. However, during the deployment phase, the original test data must be also "cooked" through DataCook to ensure normal model performance. This process grants copyright holders control over authorization during the deployment phase. The mechanism behind DataCook is by crafting anti-adversarial examples (AntiAdv), which are designed to enhance model confidence, as opposed to standard adversarial examples (Adv) that aim to confuse models. Similar to Adv, AntiAdv introduces imperceptible perturbations, ensuring that the data processed by DataCook remains easily understandable. We conducted extensive experiments on MedMNIST datasets, encompassing both 2D/3D data and the high-resolution variants. The outcomes indicate that DataCook effectively meets its objectives, preventing models trained on AntiAdv from analyzing unauthorized data effectively, without compromising the validity and accuracy of the data in legitimate scenarios. Code and data are available at https://github.com/MedMNIST/DataCook.

☆ Multi-Task Dense Prediction via Mixture of Low-Rank Experts CVPR 2024

Yuqi Yang, Peng-Tao Jiang, Qibin Hou, Hao Zhang, Jinwei Chen, Bo Li

Previous multi-task dense prediction methods based on the Mixture of Experts (MoE) have received great performance but they neglect the importance of explicitly modeling the global relations among all tasks. In this paper, we present a novel decoder-focused method for multi-task dense prediction, called Mixture-of-Low-Rank-Experts (MLoRE). To model the global task relationships, MLoRE adds a generic convolution path to the original MoE structure, where each task feature can go through this path for explicit parameter sharing. Furthermore, to control the parameters and computational cost brought by the increase in the number of experts, we take inspiration from LoRA and propose to leverage the low-rank format of a vanilla convolution in the expert network. Since the low-rank experts have fewer parameters and can be dynamically parameterized into the generic convolution, the parameters and computational cost do not change much with the increase of experts. Benefiting from this design, we increase the number of experts and its reception field to enlarge the representation capacity, facilitating multiple dense tasks learning in a unified network. Extensive experiments on the PASCAL-Context and NYUD-v2 benchmarks show that our MLoRE achieves superior performance compared to previous state-of-the-art methods on all metrics. Our code is available at https://github.com/YuqiYang213/MLoRE.

comment: Accepted at CVPR 2024

☆ Paired Diffusion: Generation of related, synthetic PET-CT-Segmentation scans using Linked Denoising Diffusion Probabilistic Models

Rowan Bradbury, Katherine A. Vallis, Bartlomiej W. Papiez

The rapid advancement of Artificial Intelligence (AI) in biomedical imaging and radiotherapy is hindered by the limited availability of large imaging data repositories. With recent research and improvements in denoising diffusion probabilistic models (DDPM), high quality synthetic medical scans are now possible. Despite this, there is currently no way of generating multiple related images, such as a corresponding ground truth which can be used to train models, so synthetic scans are often manually annotated before use. This research introduces a novel architecture that is able to generate multiple, related PET-CT-tumour mask pairs using paired networks and conditional encoders. Our approach includes innovative, time step-controlled mechanisms and a `noise-seeding' strategy to improve DDPM sampling consistency. While our model requires a modified perceptual loss function to ensure accurate feature alignment we show generation of clearly aligned synthetic images and improvement in segmentation accuracy with generated images.

comment: to be published in IEEE International Symposium on Biomedical Imaging 2024

☆ FastPerson: Enhancing Video Learning through Effective Video Summarization that Preserves Linguistic and Visual Contexts

Kazuki Kawamura, Jun Rekimoto

Quickly understanding lengthy lecture videos is essential for learners with limited time and interest in various topics to improve their learning efficiency. To this end, video summarization has been actively researched to enable users to view only important scenes from a video. However, these studies focus on either the visual or audio information of a video and extract important segments in the video. Therefore, there is a risk of missing important information when both the teacher's speech and visual information on the blackboard or slides are important, such as in a lecture video. To tackle this issue, we propose FastPerson, a video summarization approach that considers both the visual and auditory information in lecture videos. FastPerson creates summary videos by utilizing audio transcriptions along with on-screen images and text, minimizing the risk of overlooking crucial information for learners. Further, it provides a feature that allows learners to switch between the summary and original videos for each chapter of the video, enabling them to adjust the pace of learning based on their interests and level of understanding. We conducted an evaluation with 40 participants to assess the effectiveness of our method and confirmed that it reduced viewing time by 53\% at the same level of comprehension as that when using traditional video playback methods.

☆ Deep Learning for Segmentation of Cracks in High-Resolution Images of Steel Bridges

Andrii Kompanets, Gautam Pai, Remco Duits, Davide Leonetti, Bert Snijder

Automating the current bridge visual inspection practices using drones and image processing techniques is a prominent way to make these inspections more effective, robust, and less expensive. In this paper, we investigate the development of a novel deep-learning method for the detection of fatigue cracks in high-resolution images of steel bridges. First, we present a novel and challenging dataset comprising of images of cracks in steel bridges. Secondly, we integrate the ConvNext neural network with a previous state- of-the-art encoder-decoder network for crack segmentation. We study and report, the effects of the use of background patches on the network performance when applied to high-resolution images of cracks in steel bridges. Finally, we introduce a loss function that allows the use of more background patches for the training process, which yields a significant reduction in false positive rates.

☆ Invisible Gas Detection: An RGB-Thermal Cross Attention Network and A New Benchmark

Jue Wang, Yuxiang Lin, Qi Zhao, Dong Luo, Shuaibao Chen, Wei Chen, Xiaojiang Peng

The widespread use of various chemical gases in industrial processes necessitates effective measures to prevent their leakage during transportation and storage, given their high toxicity. Thermal infrared-based computer vision detection techniques provide a straightforward approach to identify gas leakage areas. However, the development of high-quality algorithms has been challenging due to the low texture in thermal images and the lack of open-source datasets. In this paper, we present the RGB-Thermal Cross Attention Network (RT-CAN), which employs an RGB-assisted two-stream network architecture to integrate texture information from RGB images and gas area information from thermal images. Additionally, to facilitate the research of invisible gas detection, we introduce Gas-DB, an extensive open-source gas detection database including about 1.3K well-annotated RGB-thermal images with eight variant collection scenes. Experimental results demonstrate that our method successfully leverages the advantages of both modalities, achieving state-of-the-art (SOTA) performance among RGB-thermal methods, surpassing single-stream SOTA models in terms of accuracy, Intersection of Union (IoU), and F2 metrics by 4.86%, 5.65%, and 4.88%, respectively. The code and data will be made available soon.

☆ Groupwise Query Specialization and Quality-Aware Multi-Assignment for Transformer-based Visual Relationship Detection CVPR 2024

Jongha Kim, Jihwan Park, Jinyoung Park, Jinyoung Kim, Sehyung Kim, Hyunwoo J. Kim

Visual Relationship Detection (VRD) has seen significant advancements with Transformer-based architectures recently. However, we identify two key limitations in a conventional label assignment for training Transformer-based VRD models, which is a process of mapping a ground-truth (GT) to a prediction. Under the conventional assignment, an unspecialized query is trained since a query is expected to detect every relation, which makes it difficult for a query to specialize in specific relations. Furthermore, a query is also insufficiently trained since a GT is assigned only to a single prediction, therefore near-correct or even correct predictions are suppressed by being assigned no relation as a GT. To address these issues, we propose Groupwise Query Specialization and Quality-Aware Multi-Assignment (SpeaQ). Groupwise Query Specialization trains a specialized query by dividing queries and relations into disjoint groups and directing a query in a specific query group solely toward relations in the corresponding relation group. Quality-Aware Multi-Assignment further facilitates the training by assigning a GT to multiple predictions that are significantly close to a GT in terms of a subject, an object, and the relation in between. Experimental results and analyses show that SpeaQ effectively trains specialized queries, which better utilize the capacity of a model, resulting in consistent performance gains with zero additional inference cost across multiple VRD models and benchmarks. Code is available at https://github.com/mlvlab/SpeaQ.

comment: CVPR 2024

☆ Panonut360: A Head and Eye Tracking Dataset for Panoramic Video ACM MM

Yutong Xu, Junhao Du, Jiahe Wang, Yuwei Ning, Sihan Zhou Yang Cao

With the rapid development and widespread application of VR/AR technology, maximizing the quality of immersive panoramic video services that match users' personal preferences and habits has become a long-standing challenge. Understanding the saliency region where users focus, based on data collected with HMDs, can promote multimedia encoding, transmission, and quality assessment. At the same time, large-scale datasets are essential for researchers and developers to explore short/long-term user behavior patterns and train AI models related to panoramic videos. However, existing panoramic video datasets often include low-frequency user head or eye movement data through short-term videos only, lacking sufficient data for analyzing users' Field of View (FoV) and generating video saliency regions. Driven by these practical factors, in this paper, we present a head and eye tracking dataset involving 50 users (25 males and 25 females) watching 15 panoramic videos. The dataset provides details on the viewport and gaze attention locations of users. Besides, we present some statistics samples extracted from the dataset. For example, the deviation between head and eye movements challenges the widely held assumption that gaze attention decreases from the center of the FoV following a Gaussian distribution. Our analysis reveals a consistent downward offset in gaze fixations relative to the FoV in experimental settings involving multiple users and videos. That's why we name the dataset Panonut, a saliency weighting shaped like a donut. Finally, we also provide a script that generates saliency distributions based on given head or eye coordinates and pre-generated saliency distribution map sets of each video from the collected eye tracking data. The dataset is available on website: https://dianvrlab.github.io/Panonut360/.

comment: 7 pages,ACM MMSys'24 accepted

☆ The Solution for the CVPR 2023 1st foundation model challenge-Track2

Haonan Xu, Yurui Huang, Sishun Pan, Zhihao Guan, Yi Xu, Yang Yang

In this paper, we propose a solution for cross-modal transportation retrieval. Due to the cross-domain problem of traffic images, we divide the problem into two sub-tasks of pedestrian retrieval and vehicle retrieval through a simple strategy. In pedestrian retrieval tasks, we use IRRA as the base model and specifically design an Attribute Classification to mine the knowledge implied by attribute labels. More importantly, We use the strategy of Inclusion Relation Matching to make the image-text pairs with inclusion relation have similar representation in the feature space. For the vehicle retrieval task, we use BLIP as the base model. Since aligning the color attributes of vehicles is challenging, we introduce attribute-based object detection techniques to add color patch blocks to vehicle images for color data augmentation. This serves as strong prior information, helping the model perform the image-text alignment. At the same time, we incorporate labeled attributes into the image-text alignment loss to learn fine-grained alignment and prevent similar images and texts from being incorrectly separated. Our approach ranked first in the final B-board test with a score of 70.9.

☆ Rotate to Scan: UNet-like Mamba with Triplet SSM Module for Medical Image Segmentation

Hao Tang, Lianglun Cheng, Guoheng Huang, Zhengguang Tan, Junhao Lu, Kaihong Wu

Image segmentation holds a vital position in the realms of diagnosis and treatment within the medical domain. Traditional convolutional neural networks (CNNs) and Transformer models have made significant advancements in this realm, but they still encounter challenges because of limited receptive field or high computing complexity. Recently, State Space Models (SSMs), particularly Mamba and its variants, have demonstrated notable performance in the field of vision. However, their feature extraction methods may not be sufficiently effective and retain some redundant structures, leaving room for parameter reduction. Motivated by previous spatial and channel attention methods, we propose Triplet Mamba-UNet. The method leverages residual VSS Blocks to extract intensive contextual features, while Triplet SSM is employed to fuse features across spatial and channel dimensions. We conducted experiments on ISIC17, ISIC18, CVC-300, CVC-ClinicDB, Kvasir-SEG, CVC-ColonDB, and Kvasir-Instrument datasets, demonstrating the superior segmentation performance of our proposed TM-UNet. Additionally, compared to the previous VM-UNet, our model achieves a one-third reduction in parameters.

★ PlainMamba: Improving Non-Hierarchical Mamba in Visual Recognition

Chenhongyi Yang, Zehui Chen, Miguel Espinosa, Linus Ericsson, Zhenyu Wang, Jiaming Liu, Elliot J. Crowley

We present PlainMamba: a simple non-hierarchical state space model (SSM) designed for general visual recognition. The recent Mamba model has shown how SSMs can be highly competitive with other architectures on sequential data and initial attempts have been made to apply it to images. In this paper, we further adapt the selective scanning process of Mamba to the visual domain, enhancing its ability to learn features from two-dimensional images by (i) a continuous 2D scanning process that improves spatial continuity by ensuring adjacency of tokens in the scanning sequence, and (ii) direction-aware updating which enables the model to discern the spatial relations of tokens by encoding directional information. Our architecture is designed to be easy to use and easy to scale, formed by stacking identical PlainMamba blocks, resulting in a model with constant width throughout all layers. The architecture is further simplified by removing the need for special tokens. We evaluate PlainMamba on a variety of visual recognition tasks including image classification, semantic segmentation, object detection, and instance segmentation. Our method achieves performance gains over previous non-hierarchical models and is competitive with hierarchical alternatives. For tasks requiring high-resolution inputs, in particular, PlainMamba requires much less computing while maintaining high performance. Code and models are available at https://github.com/ChenhongyiYang/PlainMamba

☆ AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animation

Huawei Wei, Zejun Yang, Zhisheng Wang

In this study, we propose AniPortrait, a novel framework for generating high-quality animation driven by audio and a reference portrait image. Our methodology is divided into two stages. Initially, we extract 3D intermediate representations from audio and project them into a sequence of 2D facial landmarks. Subsequently, we employ a robust diffusion model, coupled with a motion module, to convert the landmark sequence into photorealistic and temporally consistent portrait animation. Experimental results demonstrate the superiority of AniPortrait in terms of facial naturalness, pose diversity, and visual quality, thereby offering an enhanced perceptual experience. Moreover, our methodology exhibits considerable potential in terms of flexibility and controllability, which can be effectively applied in areas such as facial motion editing or face reenactment. We release code and model weights at https://github.com/scutzzj/AniPortrait

☆ Manifold-Guided Lyapunov Control with Diffusion Models

Amartya Mukherjee, Thanin Quartz, Jun Liu

This paper presents a novel approach to generating stabilizing controllers for a large class of dynamical systems using diffusion models. The core objective is to develop stabilizing control functions by identifying the closest asymptotically stable vector field relative to a predetermined manifold and adjusting the control function based on this finding. To achieve this, we employ a diffusion model trained on pairs consisting of asymptotically stable vector fields and their corresponding Lyapunov functions. Our numerical results demonstrate that this pre-trained model can achieve stabilization over previously unseen systems efficiently and rapidly, showcasing the potential of our approach in fast zero-shot control and generalizability.

comment: 14 pages

☆ Not All Similarities Are Created Equal: Leveraging Data-Driven Biases to Inform GenAI Copyright Disputes

Uri Hacohen, Adi Haviv, Shahar Sarfaty, Bruria Friedman, Niva Elkin-Koren, Roi Livni, Amit H Bermano

The advent of Generative Artificial Intelligence (GenAI) models, including GitHub Copilot, OpenAI GPT, and Stable Diffusion, has revolutionized content creation, enabling non-professionals to produce high-quality content across various domains. This transformative technology has led to a surge of synthetic content and sparked legal disputes over copyright infringement. To address these challenges, this paper introduces a novel approach that leverages the learning capacity of GenAI models for copyright legal analysis, demonstrated with GPT2 and Stable Diffusion models. Copyright law distinguishes between original expressions and generic ones (Sc\`enes \`a faire), protecting the former and permitting reproduction of the latter. However, this distinction has historically been challenging to make consistently, leading to over-protection of copyrighted works. GenAI offers an unprecedented opportunity to enhance this legal analysis by revealing shared patterns in preexisting works. We propose a data-driven approach to identify the genericity of works created by GenAI, employing "data-driven bias" to assess the genericity of expressive compositions. This approach aids in copyright scope determination by utilizing the capabilities of GenAI to identify and prioritize expressive elements and rank them according to their frequency in the model's dataset. The potential implications of measuring expressive genericity for copyright law are profound. Such scoring could assist courts in determining copyright scope during litigation, inform the registration practices of Copyright Offices, allowing registration of only highly original synthetic works, and help copyright owners signal the value of their works and facilitate fairer licensing deals. More generally, this approach offers valuable insights to policymakers grappling with adapting copyright law to the challenges posed by the era of GenAI.

comment: Presented at ACM CSLAW 2024

☆ Hierarchical Light Transformer Ensembles for Multimodal Trajectory Forecasting

Adrien Lafage, Mathieu Barbier, Gianni Franchi, David Filliat

Accurate trajectory forecasting is crucial for the performance of various systems, such as advanced driver-assistance systems and self-driving vehicles. These forecasts allow to anticipate events leading to collisions and, therefore, to mitigate them. Deep Neural Networks have excelled in motion forecasting, but issues like overconfidence and uncertainty quantification persist. Deep Ensembles address these concerns, yet applying them to multimodal distributions remains challenging. In this paper, we propose a novel approach named Hierarchical Light Transformer Ensembles (HLT-Ens), aimed at efficiently training an ensemble of Transformer architectures using a novel hierarchical loss function. HLT-Ens leverages grouped fully connected layers, inspired by grouped convolution techniques, to capture multimodal distributions, effectively. Through extensive experimentation, we demonstrate that HLT-Ens achieves state-of-the-art performance levels, offering a promising avenue for improving trajectory forecasting techniques.

☆ Predicting Perceived Gloss: Do Weak Labels Suffice?

Julia Guerrero-Viu, J. Daniel Subias, Ana Serrano, Katherine R. Storrs, Roland W. Fleming, Belen Masia, Diego Gutierrez

Estimating perceptual attributes of materials directly from images is a challenging task due to their complex, not fully-understood interactions with external factors, such as geometry and lighting. Supervised deep learning models have recently been shown to outperform traditional approaches, but rely on large datasets of human-annotated images for accurate perception predictions. Obtaining reliable annotations is a costly endeavor, aggravated by the limited ability of these models to generalise to different aspects of appearance. In this work, we show how a much smaller set of human annotations ("strong labels") can be effectively augmented with automatically derived "weak labels" in the context of learning a low-dimensional image-computable gloss metric. We evaluate three alternative weak labels for predicting human gloss perception from limited annotated data. Incorporating weak labels enhances our gloss prediction beyond the current state of the art. Moreover, it enables a substantial reduction in human annotation costs without sacrificing accuracy, whether working with rendered images or real photographs.

comment: Computer Graphics Forum (Eurographics 2024)

☆ DiffFAE: Advancing High-fidelity One-shot Facial Appearance Editing with Space-sensitive Customization and Semantic Preservation

Qilin Wang, Jiangning Zhang, Chengming Xu, Weijian Cao, Ying Tai, Yue Han, Yanhao Ge, Hong Gu, Chengjie Wang, Yanwei Fu

Facial Appearance Editing (FAE) aims to modify physical attributes, such as pose, expression and lighting, of human facial images while preserving attributes like identity and background, showing great importance in photograph. In spite of the great progress in this area, current researches generally meet three challenges: low generation fidelity, poor attribute preservation, and inefficient inference. To overcome above challenges, this paper presents DiffFAE, a one-stage and highly-efficient diffusion-based framework tailored for high-fidelity FAE. For high-fidelity query attributes transfer, we adopt Space-sensitive Physical Customization (SPC), which ensures the fidelity and generalization ability by utilizing rendering texture derived from 3D Morphable Model (3DMM). In order to preserve source attributes, we introduce the Region-responsive Semantic Composition (RSC). This module is guided to learn decoupled source-regarding features, thereby better preserving the identity and alleviating artifacts from non-facial attributes such as hair, clothes, and background. We further introduce a consistency regularization for our pipeline to enhance editing controllability by leveraging prior knowledge in the attention matrices of diffusion model. Extensive experiments demonstrate the superiority of DiffFAE over existing methods, achieving state-of-the-art performance in facial appearance editing.

☆ Exploring Dynamic Transformer for Efficient Object Tracking

Jiawen Zhu, Xin Chen, Haiwen Diao, Shuai Li, Jun-Yan He, Chenyang Li, Bin Luo, Dong Wang, Huchuan Lu

The speed-precision trade-off is a critical problem for visual object tracking which usually requires low latency and deployment on constrained resources. Existing solutions for efficient tracking mainly focus on adopting light-weight backbones or modules, which nevertheless come at the cost of a sacrifice in precision. In this paper, inspired by dynamic network routing, we propose DyTrack, a dynamic transformer framework for efficient tracking. Real-world tracking scenarios exhibit diverse levels of complexity. We argue that a simple network is sufficient for easy frames in video sequences, while more computation could be assigned to difficult ones. DyTrack automatically learns to configure proper reasoning routes for various inputs, gaining better utilization of the available computational budget. Thus, it can achieve higher performance with the same running speed. We formulate instance-specific tracking as a sequential decision problem and attach terminating branches to intermediate layers of the entire model. Especially, to fully utilize the computations, we introduce the feature recycling mechanism to reuse the outputs of predecessors. Furthermore, a target-aware self-distillation strategy is designed to enhance the discriminating capabilities of early predictions by effectively mimicking the representation pattern of the deep model. Extensive experiments on multiple benchmarks demonstrate that DyTrack achieves promising speed-precision trade-offs with only a single model. For instance, DyTrack obtains 64.9% AUC on LaSOT with a speed of 256 fps.

☆ High-Resolution Image Translation Model Based on Grayscale Redefinition

Xixian Wu, Dian Chao, Yang Yang

Image-to-image translation is a technique that focuses on transferring images from one domain to another while maintaining the essential content representations. In recent years, image-to-image translation has gained significant attention and achieved remarkable advancements due to its diverse applications in computer vision and image processing tasks. In this work, we propose an innovative method for image translation between different domains. For high-resolution image translation tasks, we use a grayscale adjustment method to achieve pixel-level translation. For other tasks, we utilize the Pix2PixHD model with a coarse-to-fine generator, multi-scale discriminator, and improved loss to enhance the image translation performance. On the other hand, to tackle the issue of sparse training data, we adopt model weight initialization from other task to optimize the performance of the current task.

☆ Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency CVPR 2024

Yingjie Xu, Bangzhen Liu, Hao Tang, Bailin Deng, Shengfeng He

We propose a voxel-based optimization framework, ReVoRF, for few-shot radiance fields that strategically address the unreliability in pseudo novel view synthesis. Our method pivots on the insight that relative depth relationships within neighboring regions are more reliable than the absolute color values in disoccluded areas. Consequently, we devise a bilateral geometric consistency loss that carefully navigates the trade-off between color fidelity and geometric accuracy in the context of depth consistency for uncertain regions. Moreover, we present a reliability-guided learning strategy to discern and utilize the variable quality across synthesized views, complemented by a reliability-aware voxel smoothing algorithm that smoothens the transition between reliable and unreliable data patches. Our approach allows for a more nuanced use of all available data, promoting enhanced learning from regions previously considered unsuitable for high-quality reconstruction. Extensive experiments across diverse datasets reveal that our approach attains significant gains in efficiency and accuracy, delivering rendering speeds of 3 FPS, 7 mins to train a $360^\circ$ scene, and a 5\% improvement in PSNR over existing few-shot methods. Code is available at https://github.com/HKCLynn/ReVoRF.

comment: CVPR 2024 final version

☆ UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection with Sparse LiDAR and Large Domain Gaps

Maciej K Wozniak, Mattias Hansson, Marko Thiel, Patric Jensfelt

In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains. Our code is open-source and will be available soon.

☆ AniArtAvatar: Animatable 3D Art Avatar from a Single Image

Shaoxu Li

We present a novel approach for generating animatable 3D-aware art avatars from a single image, with controllable facial expressions, head poses, and shoulder movements. Unlike previous reenactment methods, our approach utilizes a view-conditioned 2D diffusion model to synthesize multi-view images from a single art portrait with a neutral expression. With the generated colors and normals, we synthesize a static avatar using an SDF-based neural surface. For avatar animation, we extract control points, transfer the motion with these points, and deform the implicit canonical space. Firstly, we render the front image of the avatar, extract the 2D landmarks, and project them to the 3D space using a trained SDF network. We extract 3D driving landmarks using 3DMM and transfer the motion to the avatar landmarks. To animate the avatar pose, we manually set the body height and bound the head and torso of an avatar with two cages. The head and torso can be animated by transforming the two cages. Our approach is a one-shot pipeline that can be applied to various styles. Experiments demonstrate that our method can generate high-quality 3D art avatars with desired control over different motions.

☆ Grad-CAMO: Learning Interpretable Single-Cell Morphological Profiles from 3D Cell Painting Images

Vivek Gopalakrishnan, Jingzhe Ma, Zhiyong Xie

Despite their black-box nature, deep learning models are extensively used in image-based drug discovery to extract feature vectors from single cells in microscopy images. To better understand how these networks perform representation learning, we employ visual explainability techniques (e.g., Grad-CAM). Our analyses reveal several mechanisms by which supervised models cheat, exploiting biologically irrelevant pixels when extracting morphological features from images, such as noise in the background. This raises doubts regarding the fidelity of learned single-cell representations and their relevance when investigating downstream biological questions. To address this misalignment between researcher expectations and machine behavior, we introduce Grad-CAMO, a novel single-cell interpretability score for supervised feature extractors. Grad-CAMO measures the proportion of a model's attention that is concentrated on the cell of interest versus the background. This metric can be assessed per-cell or averaged across a validation set, offering a tool to audit individual features vectors or guide the improved design of deep learning architectures. Importantly, Grad-CAMO seamlessly integrates into existing workflows, requiring no dataset or model modifications, and is compatible with both 2D and 3D Cell Painting data. Additional results are available at https://github.com/eigenvivek/Grad-CAMO.

☆ MMVP: A Multimodal MoCap Dataset with Vision and Pressure Sensors CVPR2024

He Zhang, Shenghao Ren, Haolei Yuan, Jianhui Zhao, Fan Li, Shuangpeng Sun, Zhenghao Liang, Tao Yu, Qiu Shen, Xun Cao

Foot contact is an important cue not only for human motion capture but also for motion understanding and physically plausible motion generation. However, most of the foot-contact annotations in existing datasets are estimated by purely visual matching and distance thresholding, which results in low accuracy and coarse granularity. Even though existing multimodal datasets synergistically capture plantar pressure (foot contact) and visual signals, they are specifically designed for small-range and slow motion such as Taiji Quan and Yoga. Therefore, there is still a lack of a vision-pressure multimodal dataset with large-range and fast human motion, as well as accurate and dense foot-contact annotation. To fill this gap, we propose a Multimodal MoCap Dataset with Vision and Pressure sensors, named MMVP. MMVP provides accurate and dense plantar pressure signals synchronized with RGBD observations, which is especially useful for both plausible shape estimation, robust pose fitting without foot drifting, and accurate global translation tracking. To validate the dataset, we propose an RGBD-P SMPL fitting method and also a monocular-video-based baseline framework, VP-MoCap, for human motion capture. Experiments demonstrate that our RGBD-P SMPL Fitting results significantly outperform pure visual motion capture. Moreover, VP-MoCap outperforms SOTA methods in foot-contact and global translation estimation accuracy. We believe the configuration of the dataset and the baseline frameworks will stimulate the research in this direction and also provide a good reference for MoCap applications in various domains. Project page: https://haolyuan.github.io/MMVP-Dataset/.

comment: CVPR2024

☆ Fake or JPEG? Revealing Common Biases in Generated Image Detection Datasets

Patrick Grommelt, Louis Weiss, Franz-Josef Pfreundt, Janis Keuper

The widespread adoption of generative image models has highlighted the urgent need to detect artificial content, which is a crucial step in combating widespread manipulation and misinformation. Consequently, numerous detectors and associated datasets have emerged. However, many of these datasets inadvertently introduce undesirable biases, thereby impacting the effectiveness and evaluation of detectors. In this paper, we emphasize that many datasets for AI-generated image detection contain biases related to JPEG compression and image size. Using the GenImage dataset, we demonstrate that detectors indeed learn from these undesired factors. Furthermore, we show that removing the named biases substantially increases robustness to JPEG compression and significantly alters the cross-generator performance of evaluated detectors. Specifically, it leads to more than 11 percentage points increase in cross-generator performance for ResNet50 and Swin-T detectors on the GenImage dataset, achieving state-of-the-art results. We provide the dataset and source codes of this paper on the anonymous website: https://www.unbiased-genimage.org

☆ Dual Memory Networks: A Versatile Adaptation Approach for Vision-Language Models CVPR2024

Yabin Zhang, Wenjie Zhu, Hui Tang, Zhiyuan Ma, Kaiyang Zhou, Lei Zhang

With the emergence of pre-trained vision-language models like CLIP, how to adapt them to various downstream classification tasks has garnered significant attention in recent research. The adaptation strategies can be typically categorized into three paradigms: zero-shot adaptation, few-shot adaptation, and the recently-proposed training-free few-shot adaptation. Most existing approaches are tailored for a specific setting and can only cater to one or two of these paradigms. In this paper, we introduce a versatile adaptation approach that can effectively work under all three settings. Specifically, we propose the dual memory networks that comprise dynamic and static memory components. The static memory caches training data knowledge, enabling training-free few-shot adaptation, while the dynamic memory preserves historical test features online during the testing process, allowing for the exploration of additional data insights beyond the training set. This novel capability enhances model performance in the few-shot setting and enables model usability in the absence of training data. The two memory networks employ the same flexible memory interactive strategy, which can operate in a training-free mode and can be further enhanced by incorporating learnable projection layers. Our approach is tested across 11 datasets under the three task settings. Remarkably, in the zero-shot scenario, it outperforms existing methods by over 3\% and even shows superior results against methods utilizing external training data. Additionally, our method exhibits robust performance against natural distribution shifts. Codes are available at \url{https://github.com/YBZh/DMN}.

comment: CVPR2024; Codes are available at \url{https://github.com/YBZh/DMN}

☆ DeepMIF: Deep Monotonic Implicit Fields for Large-Scale LiDAR 3D Mapping

Kutay Yılmaz, Matthias Nießner, Anastasiia Kornilova, Alexey Artemov

Recently, significant progress has been achieved in sensing real large-scale outdoor 3D environments, particularly by using modern acquisition equipment such as LiDAR sensors. Unfortunately, they are fundamentally limited in their ability to produce dense, complete 3D scenes. To address this issue, recent learning-based methods integrate neural implicit representations and optimizable feature grids to approximate surfaces of 3D scenes. However, naively fitting samples along raw LiDAR rays leads to noisy 3D mapping results due to the nature of sparse, conflicting LiDAR measurements. Instead, in this work we depart from fitting LiDAR data exactly, instead letting the network optimize a non-metric monotonic implicit field defined in 3D space. To fit our field, we design a learning system integrating a monotonicity loss that enables optimizing neural monotonic fields and leverages recent progress in large-scale 3D mapping. Our algorithm achieves high-quality dense 3D mapping performance as captured by multiple quantitative and perceptual measures and visual results obtained for Mai City, Newer College, and KITTI benchmarks. The code of our approach will be made publicly available.

comment: 8 pages, 6 figures

☆ Practical Applications of Advanced Cloud Services and Generative AI Systems in Medical Image Analysis

Jingyu Xu, Binbin Wu, Jiaxin Huang, Yulu Gong, Yifan Zhang, Bo Liu

The medical field is one of the important fields in the application of artificial intelligence technology. With the explosive growth and diversification of medical data, as well as the continuous improvement of medical needs and challenges, artificial intelligence technology is playing an increasingly important role in the medical field. Artificial intelligence technologies represented by computer vision, natural language processing, and machine learning have been widely penetrated into diverse scenarios such as medical imaging, health management, medical information, and drug research and development, and have become an important driving force for improving the level and quality of medical services.The article explores the transformative potential of generative AI in medical imaging, emphasizing its ability to generate syntheticACM-2 data, enhance images, aid in anomaly detection, and facilitate image-to-image translation. Despite challenges like model complexity, the applications of generative models in healthcare, including Med-PaLM 2 technology, show promising results. By addressing limitations in dataset size and diversity, these models contribute to more accurate diagnoses and improved patient outcomes. However, ethical considerations and collaboration among stakeholders are essential for responsible implementation. Through experiments leveraging GANs to augment brain tumor MRI datasets, the study demonstrates how generative AI can enhance image quality and diversity, ultimately advancing medical diagnostics and patient care.

☆ A Gaze-grounded Visual Question Answering Dataset for Clarifying Ambiguous Japanese Questions LREC

Shun Inadumi, Seiya Kawano, Akishige Yuguchi, Yasutomo Kawanishi, Koichiro Yoshino

Situated conversations, which refer to visual information as visual question answering (VQA), often contain ambiguities caused by reliance on directive information. This problem is exacerbated because some languages, such as Japanese, often omit subjective or objective terms. Such ambiguities in questions are often clarified by the contexts in conversational situations, such as joint attention with a user or user gaze information. In this study, we propose the Gaze-grounded VQA dataset (GazeVQA) that clarifies ambiguous questions using gaze information by focusing on a clarification process complemented by gaze information. We also propose a method that utilizes gaze target estimation results to improve the accuracy of GazeVQA tasks. Our experimental results showed that the proposed method improved the performance in some cases of a VQA system on GazeVQA and identified some typical problems of GazeVQA tasks that need to be improved.

comment: LREC-COLING 2024

☆ WordRobe: Text-Guided Generation of Textured 3D Garments

Astitva Srivastava, Pranav Manu, Amit Raj, Varun Jampani, Avinash Sharma

In this paper, we tackle a new and challenging problem of text-driven generation of 3D garments with high-quality textures. We propose "WordRobe", a novel framework for the generation of unposed & textured 3D garment meshes from user-friendly text prompts. We achieve this by first learning a latent representation of 3D garments using a novel coarse-to-fine training strategy and a loss for latent disentanglement, promoting better latent interpolation. Subsequently, we align the garment latent space to the CLIP embedding space in a weakly supervised manner, enabling text-driven 3D garment generation and editing. For appearance modeling, we leverage the zero-shot generation capability of ControlNet to synthesize view-consistent texture maps in a single feed-forward inference step, thereby drastically decreasing the generation time as compared to existing methods. We demonstrate superior performance over current SOTAs for learning 3D garment latent space, garment interpolation, and text-driven texture synthesis, supported by quantitative evaluation and qualitative user study. The unposed 3D garment meshes generated using WordRobe can be directly fed to standard cloth simulation & animation pipelines without any post-processing.

☆ NeRF-HuGS: Improved Neural Radiance Fields in Non-static Scenes Using Heuristics-Guided Segmentation CVPR2024

Jiahao Chen, Yipeng Qin, Lingjie Liu, Jiangbo Lu, Guanbin Li

Neural Radiance Field (NeRF) has been widely recognized for its excellence in novel view synthesis and 3D scene reconstruction. However, their effectiveness is inherently tied to the assumption of static scenes, rendering them susceptible to undesirable artifacts when confronted with transient distractors such as moving objects or shadows. In this work, we propose a novel paradigm, namely "Heuristics-Guided Segmentation" (HuGS), which significantly enhances the separation of static scenes from transient distractors by harmoniously combining the strengths of hand-crafted heuristics and state-of-the-art segmentation models, thus significantly transcending the limitations of previous solutions. Furthermore, we delve into the meticulous design of heuristics, introducing a seamless fusion of Structure-from-Motion (SfM)-based heuristics and color residual heuristics, catering to a diverse range of texture profiles. Extensive experiments demonstrate the superiority and robustness of our method in mitigating transient distractors for NeRFs trained in non-static scenes. Project page: https://cnhaox.github.io/NeRF-HuGS/.

comment: To appear in CVPR2024

☆ Boosting Few-Shot Learning with Disentangled Self-Supervised Learning and Meta-Learning for Medical Image Classification

Eva Pachetti, Sotirios A. Tsaftaris, Sara Colantonio

Background and objective: Employing deep learning models in critical domains such as medical imaging poses challenges associated with the limited availability of training data. We present a strategy for improving the performance and generalization capabilities of models trained in low-data regimes. Methods: The proposed method starts with a pre-training phase, where features learned in a self-supervised learning setting are disentangled to improve the robustness of the representations for downstream tasks. We then introduce a meta-fine-tuning step, leveraging related classes between meta-training and meta-testing phases but varying the granularity level. This approach aims to enhance the model's generalization capabilities by exposing it to more challenging classification tasks during meta-training and evaluating it on easier tasks but holding greater clinical relevance during meta-testing. We demonstrate the effectiveness of the proposed approach through a series of experiments exploring several backbones, as well as diverse pre-training and fine-tuning schemes, on two distinct medical tasks, i.e., classification of prostate cancer aggressiveness from MRI data and classification of breast cancer malignity from microscopic images. Results: Our results indicate that the proposed approach consistently yields superior performance w.r.t. ablation experiments, maintaining competitiveness even when a distribution shift between training and evaluation data occurs. Conclusion: Extensive experiments demonstrate the effectiveness and wide applicability of the proposed approach. We hope that this work will add another solution to the arsenal of addressing learning issues in data-scarce imaging domains.

comment: 20 pages, 4 figures, 4 tables. Submitted to Elsevier on 25 March 2024

☆ Equipping Sketch Patches with Context-Aware Positional Encoding for Graphic Sketch Representation

Sicong Zang, Zhijun Fang

The drawing order of a sketch records how it is created stroke-by-stroke by a human being. For graphic sketch representation learning, recent studies have injected sketch drawing orders into graph edge construction by linking each patch to another in accordance to a temporal-based nearest neighboring strategy. However, such constructed graph edges may be unreliable, since a sketch could have variants of drawings. In this paper, we propose a variant-drawing-protected method by equipping sketch patches with context-aware positional encoding (PE) to make better use of drawing orders for learning graphic sketch representation. Instead of injecting sketch drawings into graph edges, we embed these sequential information into graph nodes only. More specifically, each patch embedding is equipped with a sinusoidal absolute PE to highlight the sequential position in the drawing order. And its neighboring patches, ranked by the values of self-attention scores between patch embeddings, are equipped with learnable relative PEs to restore the contextual positions within a neighborhood. During message aggregation via graph convolutional networks, a node receives both semantic contents from patch embeddings and contextual patterns from PEs by its neighbors, arriving at drawing-order-enhanced sketch representations. Experimental results indicate that our method significantly improves sketch healing and controllable sketch synthesis.

☆ Boosting Adversarial Training via Fisher-Rao Norm-based Regularization CVPR2024

Xiangyu Yin, Wenjie Ruan

Adversarial training is extensively utilized to improve the adversarial robustness of deep neural networks. Yet, mitigating the degradation of standard generalization performance in adversarial-trained models remains an open problem. This paper attempts to resolve this issue through the lens of model complexity. First, We leverage the Fisher-Rao norm, a geometrically invariant metric for model complexity, to establish the non-trivial bounds of the Cross-Entropy Loss-based Rademacher complexity for a ReLU-activated Multi-Layer Perceptron. Then we generalize a complexity-related variable, which is sensitive to the changes in model width and the trade-off factors in adversarial training. Moreover, intensive empirical evidence validates that this variable highly correlates with the generalization gap of Cross-Entropy loss between adversarial-trained and standard-trained models, especially during the initial and final phases of the training process. Building upon this observation, we propose a novel regularization framework, called Logit-Oriented Adversarial Training (LOAT), which can mitigate the trade-off between robustness and accuracy while imposing only a negligible increase in computational overhead. Our extensive experiments demonstrate that the proposed regularization strategy can boost the performance of the prevalent adversarial training algorithms, including PGD-AT, TRADES, TRADES (LSE), MART, and DM-AT, across various network architectures. Our code will be available at https://github.com/TrustAI/LOAT.

comment: This paper has been accepted to CVPR2024

☆ Random-coupled Neural Network

Haoran Liu, Mingzhe Liu, Peng Li, Jiahui Wu, Xin Jiang, Zhuo Zuo, Bingqi Liu

Improving the efficiency of current neural networks and modeling them in biological neural systems have become popular research directions in recent years. Pulse-coupled neural network (PCNN) is a well applicated model for imitating the computation characteristics of the human brain in computer vision and neural network fields. However, differences between the PCNN and biological neural systems remain: limited neural connection, high computational cost, and lack of stochastic property. In this study, random-coupled neural network (RCNN) is proposed. It overcomes these difficulties in PCNN's neuromorphic computing via a random inactivation process. This process randomly closes some neural connections in the RCNN model, realized by the random inactivation weight matrix of link input. This releases the computational burden of PCNN, making it affordable to achieve vast neural connections. Furthermore, the image and video processing mechanisms of RCNN are researched. It encodes constant stimuli as periodic spike trains and periodic stimuli as chaotic spike trains, the same as biological neural information encoding characteristics. Finally, the RCNN is applicated to image segmentation, fusion, and pulse shape discrimination subtasks. It is demonstrated to be robust, efficient, and highly anti-noised, with outstanding performance in all applications mentioned above.

☆ DS-AL: A Dual-Stream Analytic Learning for Exemplar-Free Class-Incremental Learning AAAI 2024

Huiping Zhuang, Run He, Kai Tong, Ziqian Zeng, Cen Chen, Zhiping Lin

Class-incremental learning (CIL) under an exemplar-free constraint has presented a significant challenge. Existing methods adhering to this constraint are prone to catastrophic forgetting, far more so than replay-based techniques that retain access to past samples. In this paper, to solve the exemplar-free CIL problem, we propose a Dual-Stream Analytic Learning (DS-AL) approach. The DS-AL contains a main stream offering an analytical (i.e., closed-form) linear solution, and a compensation stream improving the inherent under-fitting limitation due to adopting linear mapping. The main stream redefines the CIL problem into a Concatenated Recursive Least Squares (C-RLS) task, allowing an equivalence between the CIL and its joint-learning counterpart. The compensation stream is governed by a Dual-Activation Compensation (DAC) module. This module re-activates the embedding with a different activation function from the main stream one, and seeks fitting compensation by projecting the embedding to the null space of the main stream's linear mapping. Empirical results demonstrate that the DS-AL, despite being an exemplar-free technique, delivers performance comparable with or better than that of replay-based methods across various datasets, including CIFAR-100, ImageNet-100 and ImageNet-Full. Additionally, the C-RLS' equivalent property allows the DS-AL to execute CIL in a phase-invariant manner. This is evidenced by a never-before-seen 500-phase CIL ImageNet task, which performs on a level identical to a 5-phase one. Our codes are available at https://github.com/ZHUANGHP/Analytic-continual-learning.

comment: Accepted in AAAI 2024

☆ SeNM-VAE: Semi-Supervised Noise Modeling with Hierarchical Variational Autoencoder

Dihan Zheng, Yihang Zou, Xiaowen Zhang, Chenglong Bao

The data bottleneck has emerged as a fundamental challenge in learning based image restoration methods. Researchers have attempted to generate synthesized training data using paired or unpaired samples to address this challenge. This study proposes SeNM-VAE, a semi-supervised noise modeling method that leverages both paired and unpaired datasets to generate realistic degraded data. Our approach is based on modeling the conditional distribution of degraded and clean images with a specially designed graphical model. Under the variational inference framework, we develop an objective function for handling both paired and unpaired data. We employ our method to generate paired training samples for real-world image denoising and super-resolution tasks. Our approach excels in the quality of synthetic degraded images compared to other unpaired and paired noise modeling methods. Furthermore, our approach demonstrates remarkable performance in downstream image restoration tasks, even with limited paired data. With more paired data, our method achieves the best performance on the SIDD dataset.

☆ Sharing the Cost of Success: A Game for Evaluating and Learning Collaborative Multi-Agent Instruction Giving and Following Policies LREC

Philipp Sadler, Sherzod Hakimov, David Schlangen

In collaborative goal-oriented settings, the participants are not only interested in achieving a successful outcome, but do also implicitly negotiate the effort they put into the interaction (by adapting to each other). In this work, we propose a challenging interactive reference game that requires two players to coordinate on vision and language observations. The learning signal in this game is a score (given after playing) that takes into account the achieved goal and the players' assumed efforts during the interaction. We show that a standard Proximal Policy Optimization (PPO) setup achieves a high success rate when bootstrapped with heuristic partner behaviors that implement insights from the analysis of human-human interactions. And we find that a pairing of neural partners indeed reduces the measured joint effort when playing together repeatedly. However, we observe that in comparison to a reasonable heuristic pairing there is still room for improvement -- which invites further research in the direction of cost-sharing in collaborative interactions.

comment: 9 pages, Accepted at LREC-COLING 2024

☆ Dr.Hair: Reconstructing Scalp-Connected Hair Strands without Pre-training via Differentiable Rendering of Line Segments CVPR 2024

Yusuke Takimoto, Hikari Takehara, Hiroyuki Sato, Zihao Zhu, Bo Zheng

In the film and gaming industries, achieving a realistic hair appearance typically involves the use of strands originating from the scalp. However, reconstructing these strands from observed surface images of hair presents significant challenges. The difficulty in acquiring Ground Truth (GT) data has led state-of-the-art learning-based methods to rely on pre-training with manually prepared synthetic CG data. This process is not only labor-intensive and costly but also introduces complications due to the domain gap when compared to real-world data. In this study, we propose an optimization-based approach that eliminates the need for pre-training. Our method represents hair strands as line segments growing from the scalp and optimizes them using a novel differentiable rendering algorithm. To robustly optimize a substantial number of slender explicit geometries, we introduce 3D orientation estimation utilizing global optimization, strand initialization based on Laplace's equation, and reparameterization that leverages geometric connectivity and spatial proximity. Unlike existing optimization-based methods, our method is capable of reconstructing internal hair flow in an absolute direction. Our method exhibits robust and accurate inverse rendering, surpassing the quality of existing methods and significantly improving processing speed.

comment: CVPR 2024

☆ DiffGaze: A Diffusion Model for Continuous Gaze Sequence Generation on 360° Images

Chuhan Jiao, Yao Wang, Guanhua Zhang, Mihai Bâce, Zhiming Hu, Andreas Bulling

We present DiffGaze, a novel method for generating realistic and diverse continuous human gaze sequences on 360{\deg} images based on a conditional score-based denoising diffusion model. Generating human gaze on 360{\deg} images is important for various human-computer interaction and computer graphics applications, e.g. for creating large-scale eye tracking datasets or for realistic animation of virtual humans. However, existing methods are limited to predicting discrete fixation sequences or aggregated saliency maps, thereby neglecting crucial parts of natural gaze behaviour. Our method uses features extracted from 360{\deg} images as condition and uses two transformers to model the temporal and spatial dependencies of continuous human gaze. We evaluate DiffGaze on two 360{\deg} image benchmarks for gaze sequence generation as well as scanpath prediction and saliency prediction. Our evaluations show that DiffGaze outperforms state-of-the-art methods on all tasks on both benchmarks. We also report a 21-participant user study showing that our method generates gaze sequences that are indistinguishable from real human sequences.

☆ LaRE^2: Latent Reconstruction Error Based Method for Diffusion-Generated Image Detection CVPR 2024

Yunpeng Luo, Junlong Du, Ke Yan, Shouhong Ding

The evolution of Diffusion Models has dramatically improved image generation quality, making it increasingly difficult to differentiate between real and generated images. This development, while impressive, also raises significant privacy and security concerns. In response to this, we propose a novel Latent REconstruction error guided feature REfinement method (LaRE^2) for detecting the diffusion-generated images. We come up with the Latent Reconstruction Error (LaRE), the first reconstruction-error based feature in the latent space for generated image detection. LaRE surpasses existing methods in terms of feature extraction efficiency while preserving crucial cues required to differentiate between the real and the fake. To exploit LaRE, we propose an Error-Guided feature REfinement module (EGRE), which can refine the image feature guided by LaRE to enhance the discriminativeness of the feature. Our EGRE utilizes an align-then-refine mechanism, which effectively refines the image feature for generated-image detection from both spatial and channel perspectives. Extensive experiments on the large-scale GenImage benchmark demonstrate the superiority of our LaRE^2, which surpasses the best SoTA method by up to 11.9%/12.1% average ACC/AP across 8 different image generators. LaRE also surpasses existing methods in terms of feature extraction cost, delivering an impressive speed enhancement of 8 times.

comment: CVPR 2024

☆ Building Bridges across Spatial and Temporal Resolutions: Reference-Based Super-Resolution via Change Priors and Conditional Diffusion Model CVPR2024

Runmin Dong, Shuai Yuan, Bin Luo, Mengxuan Chen, Jinxiao Zhang, Lixian Zhang, Weijia Li, Juepeng Zheng, Haohuan Fu

Reference-based super-resolution (RefSR) has the potential to build bridges across spatial and temporal resolutions of remote sensing images. However, existing RefSR methods are limited by the faithfulness of content reconstruction and the effectiveness of texture transfer in large scaling factors. Conditional diffusion models have opened up new opportunities for generating realistic high-resolution images, but effectively utilizing reference images within these models remains an area for further exploration. Furthermore, content fidelity is difficult to guarantee in areas without relevant reference information. To solve these issues, we propose a change-aware diffusion model named Ref-Diff for RefSR, using the land cover change priors to guide the denoising process explicitly. Specifically, we inject the priors into the denoising model to improve the utilization of reference information in unchanged areas and regulate the reconstruction of semantically relevant content in changed areas. With this powerful guidance, we decouple the semantics-guided denoising and reference texture-guided denoising processes to improve the model performance. Extensive experiments demonstrate the superior effectiveness and robustness of the proposed method compared with state-of-the-art RefSR methods in both quantitative and qualitative evaluations. The code and data are available at https://github.com/dongrunmin/RefDiff.

comment: Accepted by CVPR2024

☆ Chain of Compression: A Systematic Approach to Combinationally Compress Convolutional Neural Networks

Yingtao Shen, Minqing Sun, Jie Zhao, An Zou

Convolutional neural networks (CNNs) have achieved significant popularity, but their computational and memory intensity poses challenges for resource-constrained computing systems, particularly with the prerequisite of real-time performance. To release this burden, model compression has become an important research focus. Many approaches like quantization, pruning, early exit, and knowledge distillation have demonstrated the effect of reducing redundancy in neural networks. Upon closer examination, it becomes apparent that each approach capitalizes on its unique features to compress the neural network, and they can also exhibit complementary behavior when combined. To explore the interactions and reap the benefits from the complementary features, we propose the Chain of Compression, which works on the combinational sequence to apply these common techniques to compress the neural network. Validated on the image-based regression and classification networks across different data sets, our proposed Chain of Compression can significantly compress the computation cost by 100-1000 times with ignorable accuracy loss compared with the baseline model.

comment: 10 pages, 15 figures

☆ Integrating Mamba Sequence Model and Hierarchical Upsampling Network for Accurate Semantic Segmentation of Multiple Sclerosis Legion

Kazi Shahriar Sanjid, Md. Tanzim Hossain, Md. Shakib Shahariar Junayed, Dr. Mohammad Monir Uddin

Integrating components from convolutional neural networks and state space models in medical image segmentation presents a compelling approach to enhance accuracy and efficiency. We introduce Mamba HUNet, a novel architecture tailored for robust and efficient segmentation tasks. Leveraging strengths from Mamba UNet and the lighter version of Hierarchical Upsampling Network (HUNet), Mamba HUNet combines convolutional neural networks local feature extraction power with state space models long range dependency modeling capabilities. We first converted HUNet into a lighter version, maintaining performance parity and then integrated this lighter HUNet into Mamba HUNet, further enhancing its efficiency. The architecture partitions input grayscale images into patches, transforming them into 1D sequences for processing efficiency akin to Vision Transformers and Mamba models. Through Visual State Space blocks and patch merging layers, hierarchical features are extracted while preserving spatial information. Experimental results on publicly available Magnetic Resonance Imaging scans, notably in Multiple Sclerosis lesion segmentation, demonstrate Mamba HUNet's effectiveness across diverse segmentation tasks. The model's robustness and flexibility underscore its potential in handling complex anatomical structures. These findings establish Mamba HUNet as a promising solution in advancing medical image segmentation, with implications for improving clinical decision making processes.

comment: 13 pages

☆ Test-time Adaptation Meets Image Enhancement: Improving Accuracy via Uncertainty-aware Logit Switching IJCNN2024

Shohei Enomoto, Naoya Hasegawa, Kazuki Adachi, Taku Sasaki, Shin'ya Yamaguchi, Satoshi Suzuki, Takeharu Eda

Deep neural networks have achieved remarkable success in a variety of computer vision applications. However, there is a problem of degrading accuracy when the data distribution shifts between training and testing. As a solution of this problem, Test-time Adaptation~(TTA) has been well studied because of its practicality. Although TTA methods increase accuracy under distribution shift by updating the model at test time, using high-uncertainty predictions is known to degrade accuracy. Since the input image is the root of the distribution shift, we incorporate a new perspective on enhancing the input image into TTA methods to reduce the prediction's uncertainty. We hypothesize that enhancing the input image reduces prediction's uncertainty and increase the accuracy of TTA methods. On the basis of our hypothesis, we propose a novel method: Test-time Enhancer and Classifier Adaptation~(TECA). In TECA, the classification model is combined with the image enhancement model that transforms input images into recognition-friendly ones, and these models are updated by existing TTA methods. Furthermore, we found that the prediction from the enhanced image does not always have lower uncertainty than the prediction from the original image. Thus, we propose logit switching, which compares the uncertainty measure of these predictions and outputs the lower one. In our experiments, we evaluate TECA with various TTA methods and show that TECA reduces prediction's uncertainty and increases accuracy of TTA methods despite having no hyperparameters and little parameter overhead.

comment: Accepted to IJCNN2024

☆ InterHandGen: Two-Hand Interaction Generation via Cascaded Reverse Diffusion CVPR 2024

Jihyun Lee, Shunsuke Saito, Giljoo Nam, Minhyuk Sung, Tae-Kyun Kim

We present InterHandGen, a novel framework that learns the generative prior of two-hand interaction. Sampling from our model yields plausible and diverse two-hand shapes in close interaction with or without an object. Our prior can be incorporated into any optimization or learning methods to reduce ambiguity in an ill-posed setup. Our key observation is that directly modeling the joint distribution of multiple instances imposes high learning complexity due to its combinatorial nature. Thus, we propose to decompose the modeling of joint distribution into the modeling of factored unconditional and conditional single instance distribution. In particular, we introduce a diffusion model that learns the single-hand distribution unconditional and conditional to another hand via conditioning dropout. For sampling, we combine anti-penetration and classifier-free guidance to enable plausible generation. Furthermore, we establish the rigorous evaluation protocol of two-hand synthesis, where our method significantly outperforms baseline generative models in terms of plausibility and diversity. We also demonstrate that our diffusion prior can boost the performance of two-hand reconstruction from monocular in-the-wild images, achieving new state-of-the-art accuracy.

comment: Accepted to CVPR 2024, project page: https://jyunlee.github.io/projects/interhandgen/

☆ Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge CVPR 2024

Dongjin Kim, Sung Jin Um, Sangmin Lee, Jung Uk Kim

The goal of the multi-sound source localization task is to localize sound sources from the mixture individually. While recent multi-sound source localization methods have shown improved performance, they face challenges due to their reliance on prior information about the number of objects to be separated. In this paper, to overcome this limitation, we present a novel multi-sound source localization method that can perform localization without prior knowledge of the number of sound sources. To achieve this goal, we propose an iterative object identification (IOI) module, which can recognize sound-making objects in an iterative manner. After finding the regions of sound-making objects, we devise object similarity-aware clustering (OSC) loss to guide the IOI module to effectively combine regions of the same object but also distinguish between different objects and backgrounds. It enables our method to perform accurate localization of sound-making objects without any prior knowledge. Extensive experimental results on the MUSIC and VGGSound benchmarks show the significant performance improvements of the proposed method over the existing methods for both single and multi-source. Our code is available at: https://github.com/VisualAIKHU/NoPrior_MultiSSL

comment: Accepted at CVPR 2024

☆ Neural Clustering based Visual Representation Learning CVPR 2024

Guikun Chen, Xia Li, Yi Yang, Wenguan Wang

We investigate a fundamental aspect of machine vision: the measurement of features, by revisiting clustering, one of the most classic approaches in machine learning and data analysis. Existing visual feature extractors, including ConvNets, ViTs, and MLPs, represent an image as rectangular regions. Though prevalent, such a grid-style paradigm is built upon engineering practice and lacks explicit modeling of data distribution. In this work, we propose feature extraction with clustering (FEC), a conceptually elegant yet surprisingly ad-hoc interpretable neural clustering framework, which views feature extraction as a process of selecting representatives from data and thus automatically captures the underlying data distribution. Given an image, FEC alternates between grouping pixels into individual clusters to abstract representatives and updating the deep features of pixels with current representatives. Such an iterative working mechanism is implemented in the form of several neural layers and the final representatives can be used for downstream tasks. The cluster assignments across layers, which can be viewed and inspected by humans, make the forward process of FEC fully transparent and empower it with promising ad-hoc interpretability. Extensive experiments on various visual recognition models and tasks verify the effectiveness, generality, and interpretability of FEC. We expect this work will provoke a rethink of the current de facto grid-style paradigm.

comment: CVPR 2024. Code: https://github.com/guikunchen/FEC/

☆ SSF3D: Strict Semi-Supervised 3D Object Detection with Switching Filter

Songbur Wong

SSF3D modified the semi-supervised 3D object detection (SS3DOD) framework, which designed specifically for point cloud data. Leveraging the characteristics of non-coincidence and weak correlation of target objects in point cloud, we adopt a strategy of retaining only the truth-determining pseudo labels and trimming the other fuzzy labels with points, instead of pursuing a balance between the quantity and quality of pseudo labels. Besides, we notice that changing the filter will make the model meet different distributed targets, which is beneficial to break the training bottleneck. Two mechanism are introduced to achieve above ideas: strict threshold and filter switching. The experiments are conducted to analyze the effectiveness of above approaches and their impact on the overall performance of the system. Evaluating on the KITTI dataset, SSF3D exhibits superior performance compared to the current state-of-the-art methods. The code will be released here.

☆ Decoupled Pseudo-labeling for Semi-Supervised Monocular 3D Object Detection CVPR2024

Jiacheng Zhang, Jiaming Li, Xiangru Lin, Wei Zhang, Xiao Tan, Junyu Han, Errui Ding, Jingdong Wang, Guanbin Li

We delve into pseudo-labeling for semi-supervised monocular 3D object detection (SSM3OD) and discover two primary issues: a misalignment between the prediction quality of 3D and 2D attributes and the tendency of depth supervision derived from pseudo-labels to be noisy, leading to significant optimization conflicts with other reliable forms of supervision. We introduce a novel decoupled pseudo-labeling (DPL) approach for SSM3OD. Our approach features a Decoupled Pseudo-label Generation (DPG) module, designed to efficiently generate pseudo-labels by separately processing 2D and 3D attributes. This module incorporates a unique homography-based method for identifying dependable pseudo-labels in BEV space, specifically for 3D attributes. Additionally, we present a DepthGradient Projection (DGP) module to mitigate optimization conflicts caused by noisy depth supervision of pseudo-labels, effectively decoupling the depth gradient and removing conflicting gradients. This dual decoupling strategy-at both the pseudo-label generation and gradient levels-significantly improves the utilization of pseudo-labels in SSM3OD. Our comprehensive experiments on the KITTI benchmark demonstrate the superiority of our method over existing approaches.

comment: To appear in CVPR2024

☆ Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance

Donghoon Ahn, Hyoungwon Cho, Jaewon Min, Wooseok Jang, Jungwoo Kim, SeonHwa Kim, Hyun Hee Park, Kyong Hwan Jin, Seungryong Kim

Recent studies have demonstrated that diffusion models are capable of generating high-quality samples, but their quality heavily depends on sampling guidance techniques, such as classifier guidance (CG) and classifier-free guidance (CFG). These techniques are often not applicable in unconditional generation or in various downstream tasks such as image restoration. In this paper, we propose a novel sampling guidance, called Perturbed-Attention Guidance (PAG), which improves diffusion sample quality across both unconditional and conditional settings, achieving this without requiring additional training or the integration of external modules. PAG is designed to progressively enhance the structure of samples throughout the denoising process. It involves generating intermediate samples with degraded structure by substituting selected self-attention maps in diffusion U-Net with an identity matrix, by considering the self-attention mechanisms' ability to capture structural information, and guiding the denoising process away from these degraded samples. In both ADM and Stable Diffusion, PAG surprisingly improves sample quality in conditional and even unconditional scenarios. Moreover, PAG significantly improves the baseline performance in various downstream tasks where existing guidances such as CG or CFG cannot be fully utilized, including ControlNet with empty prompts and image restoration such as inpainting and deblurring.

comment: Project page is available at https://ku-cvlab.github.io/Perturbed-Attention-Guidance

☆ AIDE: An Automatic Data Engine for Object Detection in Autonomous Driving CVPR-2024

Mingfu Liang, Jong-Chyi Su, Samuel Schulter, Sparsh Garg, Shiyu Zhao, Ying Wu, Manmohan Chandraker

Autonomous vehicle (AV) systems rely on robust perception models as a cornerstone of safety assurance. However, objects encountered on the road exhibit a long-tailed distribution, with rare or unseen categories posing challenges to a deployed perception model. This necessitates an expensive process of continuously curating and annotating data with significant human effort. We propose to leverage recent advances in vision-language and large language models to design an Automatic Data Engine (AIDE) that automatically identifies issues, efficiently curates data, improves the model through auto-labeling, and verifies the model through generation of diverse scenarios. This process operates iteratively, allowing for continuous self-improvement of the model. We further establish a benchmark for open-world detection on AV datasets to comprehensively evaluate various learning paradigms, demonstrating our method's superior performance at a reduced cost.

comment: Accepted by CVPR-2024

☆ CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware Visual Prompt Tuning

Ziyang Gong, Fuhao Li, Yupeng Deng, Deblina Bhattacharjee, Xiangwei Zhu, Zhenming Ji

Unsupervised Domain Adaptation (UDA) aims to adapt models from labeled source domains to unlabeled target domains. When adapting to adverse scenes, existing UDA methods fail to perform well due to the lack of instructions, leading their models to overlook discrepancies within all adverse scenes. To tackle this, we propose CoDA which instructs models to distinguish, focus, and learn from these discrepancies at scene and image levels. Specifically, CoDA consists of a Chain-of-Domain (CoD) strategy and a Severity-Aware Visual Prompt Tuning (SAVPT) mechanism. CoD focuses on scene-level instructions to divide all adverse scenes into easy and hard scenes, guiding models to adapt from source to easy domains with easy scene images, and then to hard domains with hard scene images, thereby laying a solid foundation for whole adaptations. Building upon this foundation, we employ SAVPT to dive into more detailed image-level instructions to boost performance. SAVPT features a novel metric Severity that divides all adverse scene images into low-severity and high-severity images. Then Severity directs visual prompts and adapters, instructing models to concentrate on unified severity features instead of scene-specific features, without adding complexity to the model architecture. CoDA achieves SOTA performances on widely-used benchmarks under all adverse scenes. Notably, CoDA outperforms the existing ones by 4.6%, and 10.3% mIoU on the Foggy Driving, and Foggy Zurich benchmarks, respectively. Our code is available at https://github.com/Cuzyoung/CoDA

☆ Activity-Biometrics: Person Identification from Daily Activities CVPR 2024

Shehreen Azad, Yogesh Singh Rawat

In this work, we study a novel problem which focuses on person identification while performing daily activities. Learning biometric features from RGB videos is challenging due to spatio-temporal complexity and presence of appearance biases such as clothing color and background. We propose ABNet, a novel framework which leverages disentanglement of biometric and non-biometric features to perform effective person identification from daily activities. ABNet relies on a bias-less teacher to learn biometric features from RGB videos and explicitly disentangle non-biometric features with the help of biometric distortion. In addition, ABNet also exploits activity prior for biometrics which is enabled by joint biometric and activity learning. We perform comprehensive evaluation of the proposed approach across five different datasets which are derived from existing activity recognition benchmarks. Furthermore, we extensively compare ABNet with existing works in person identification and demonstrate its effectiveness for activity-based biometrics across all five datasets. The code and dataset can be accessed at: \url{https://github.com/sacrcv/Activity-Biometrics/}

comment: CVPR 2024 Main conference

☆ TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos

Yufu Wang, Ziyun Wang, Lingjie Liu, Kostas Daniilidis

We propose TRAM, a two-stage method to reconstruct a human's global trajectory and motion from in-the-wild videos. TRAM robustifies SLAM to recover the camera motion in the presence of dynamic humans and uses the scene background to derive the motion scale. Using the recovered camera as a metric-scale reference frame, we introduce a video transformer model (VIMO) to regress the kinematic body motion of a human. By composing the two motions, we achieve accurate recovery of 3D humans in the world space, reducing global motion errors by 60% from prior work. https://yufu-wang.github.io/tram4d/

comment: The project website: https://yufu-wang.github.io/tram4d/

☆ The Solution for the ICCV 2023 1st Scientific Figure Captioning Challenge

Dian Chao, Xin Song, Shupeng Zhong, Boyuan Wang, Xiangyu Wu, Chen Zhu, Yang Yang

In this paper, we propose a solution for improving the quality of captions generated for figures in papers. We adopt the approach of summarizing the textual content in the paper to generate image captions. Throughout our study, we encounter discrepancies in the OCR information provided in the official dataset. To rectify this, we employ the PaddleOCR toolkit to extract OCR information from all images. Moreover, we observe that certain textual content in the official paper pertains to images that are not relevant for captioning, thereby introducing noise during caption generation. To mitigate this issue, we leverage LLaMA to extract image-specific information by querying the textual content based on image mentions, effectively filtering out extraneous information. Additionally, we recognize a discrepancy between the primary use of maximum likelihood estimation during text generation and the evaluation metrics such as ROUGE employed to assess the quality of generated captions. To bridge this gap, we integrate the BRIO model framework, enabling a more coherent alignment between the generation and evaluation processes. Our approach ranked first in the final test with a score of 4.49.

☆ OVER-NAV: Elevating Iterative Vision-and-Language Navigation with Open-Vocabulary Detection and StructurEd Representation CVPR 2024

Ganlong Zhao, Guanbin Li, Weikai Chen, Yizhou Yu

Recent advances in Iterative Vision-and-Language Navigation (IVLN) introduce a more meaningful and practical paradigm of VLN by maintaining the agent's memory across tours of scenes. Although the long-term memory aligns better with the persistent nature of the VLN task, it poses more challenges on how to utilize the highly unstructured navigation memory with extremely sparse supervision. Towards this end, we propose OVER-NAV, which aims to go over and beyond the current arts of IVLN techniques. In particular, we propose to incorporate LLMs and open-vocabulary detectors to distill key information and establish correspondence between multi-modal signals. Such a mechanism introduces reliable cross-modal supervision and enables on-the-fly generalization to unseen scenes without the need of extra annotation and re-training. To fully exploit the interpreted navigation data, we further introduce a structured representation, coded Omnigraph, to effectively integrate multi-modal information along the tour. Accompanied with a novel omnigraph fusion mechanism, OVER-NAV is able to extract the most relevant knowledge from omnigraph for a more accurate navigating action. In addition, OVER-NAV seamlessly supports both discrete and continuous environments under a unified framework. We demonstrate the superiority of OVER-NAV in extensive experiments.

comment: Accepted by CVPR 2024

☆ Labeling subtypes in a Parkinson's Cohort using Multifeatures in MRI - Integrating Grey and White Matter Information

Tanmayee Samantaray, Jitender Saini, Pramod Kumar Pal, Bithiah Grace Jaganathan, Vijaya V Saradhi, Gupta CN

Thresholding of networks has long posed a challenge in brain connectivity analysis. Weighted networks are typically binarized using threshold measures to facilitate network analysis. Previous studies on MRI-based brain networks have predominantly utilized density or sparsity-based thresholding techniques, optimized within specific ranges derived from network metrics such as path length, clustering coefficient, and small-world index. Thus, determination of a single threshold value for facilitating comparative analysis of networks remains elusive. To address this, our study introduces Mutual K-Nearest Neighbor (MKNN)-based thresholding for brain network analysis. Here, nearest neighbor selection is based on the highest correlation between features of brain regions. Construction of brain networks was accomplished by computing Pearson correlations between grey matter volume and white matter volume for each pair of brain regions. Structural MRI data from 180 Parkinsons patients and 70 controls from the NIMHANS, India were analyzed. Subtypes within Parkinsons disease were identified based on grey and white matter volume atrophy using source-based morphometric decomposition. The loading coefficients were correlated with clinical features to discern clinical relationship with the deciphered subtypes. Our data-mining approach revealed: Subtype A (N = 51, intermediate type), Subtype B (N = 57, mild-severe type with mild motor symptoms), and Subtype AB (N = 36, most-severe type with predominance in motor impairment). Subtype-specific weighted matrices were binarized using MKNN-based thresholding for brain network analysis. Permutation tests on network metrics of resulting bipartite graphs demonstrated significant group differences in betweenness centrality and participation coefficient. The identified hubs were specific to each subtype, with some hubs conserved across different subtypes.

comment: 31 pages, 10 figures, 3 tables

☆ Staircase Localization for Autonomous Exploration in Urban Environments

Jinrae Kim, Sunggoo Jung, Sung-Kyun Kim, Youdan Kim, Ali-akbar Agha-mohammadi

A staircase localization method is proposed for robots to explore urban environments autonomously. The proposed method employs a modular design in the form of a cascade pipeline consisting of three modules of stair detection, line segment detection, and stair localization modules. The stair detection module utilizes an object detection algorithm based on deep learning to generate a region of interest (ROI). From the ROI, line segment features are extracted using a deep line segment detection algorithm. The extracted line segments are used to localize a staircase in terms of position, orientation, and stair direction. The stair detection and localization are performed only with a single RGB-D camera. Each component of the proposed pipeline does not need to be designed particularly for staircases, which makes it easy to maintain the whole pipeline and replace each component with state-of-the-art deep learning detection techniques. The results of real-world experiments show that the proposed method can perform accurate stair detection and localization during autonomous exploration for various structured and unstructured upstairs and downstairs with shadows, dirt, and occlusions by artificial and natural objects.

comment: 9 pages, 10 figures

☆ Accuracy enhancement method for speech emotion recognition from spectrogram using temporal frequency correlation and positional information learning through knowledge transfer

Jeong-Yoon Kim, Seung-Ho Lee

In this paper, we propose a method to improve the accuracy of speech emotion recognition (SER) by using vision transformer (ViT) to attend to the correlation of frequency (y-axis) with time (x-axis) in spectrogram and transferring positional information between ViT through knowledge transfer. The proposed method has the following originality i) We use vertically segmented patches of log-Mel spectrogram to analyze the correlation of frequencies over time. This type of patch allows us to correlate the most relevant frequencies for a particular emotion with the time they were uttered. ii) We propose the use of image coordinate encoding, an absolute positional encoding suitable for ViT. By normalizing the x, y coordinates of the image to -1 to 1 and concatenating them to the image, we can effectively provide valid absolute positional information for ViT. iii) Through feature map matching, the locality and location information of the teacher network is effectively transmitted to the student network. Teacher network is a ViT that contains locality of convolutional stem and absolute position information through image coordinate encoding, and student network is a structure that lacks positional encoding in the basic ViT structure. In feature map matching stage, we train through the mean absolute error (L1 loss) to minimize the difference between the feature maps of the two networks. To validate the proposed method, three emotion datasets (SAVEE, EmoDB, and CREMA-D) consisting of speech were converted into log-Mel spectrograms for comparison experiments. The experimental results show that the proposed method significantly outperforms the state-of-the-art methods in terms of weighted accuracy while requiring significantly fewer floating point operations (FLOPs). Overall, the proposed method offers an promising solution for SER by providing improved efficiency and performance.

☆ Tracing and segmentation of molecular patterns in 3-dimensional cryo-et/em density maps through algorithmic image processing and deep learning-based techniques

Salim Sazzed

Understanding the structures of biological macromolecules is highly important as they are closely associated with cellular functionalities. Comprehending the precise organization actin filaments is crucial because they form the dynamic cytoskeleton, which offers structural support to cells and connects the cell's interior with its surroundings. However, determining the precise organization of actin filaments is challenging due to the poor quality of cryo-electron tomography (cryo-ET) images, which suffer from low signal-to-noise (SNR) ratios and the presence of missing wedge, as well as diverse shape characteristics of actin filaments. To address these formidable challenges, the primary component of this dissertation focuses on developing sophisticated computational techniques for tracing actin filaments. In particular, three novel methodologies have been developed: i) BundleTrac, for tracing bundle-like actin filaments found in Stereocilium, ii) Spaghetti Tracer, for tracing filaments that move individually with loosely cohesive movements, and iii) Struwwel Tracer, for tracing randomly orientated actin filaments in the actin network. The second component of the dissertation introduces a convolutional neural network (CNN) based segmentation model to determine the location of protein secondary structures, such as helices and beta-sheets, in medium-resolution (5-10 Angstrom) 3-dimensional cryo-electron microscopy (cryo-EM) images. This methodology later evolved into a tool named DeepSSETracer. The final component of the dissertation presents a novel algorithm, cylindrical fit measure, to estimate image structure match at helix regions in medium-resolution cryo-EM images. Overall, my dissertation has made significant contributions to addressing critical research challenges in structural biology by introducing various computational methods and tools.

♻ ☆ VMRNN: Integrating Vision Mamba and LSTM for Efficient and Accurate Spatiotemporal Forecasting

Yujin Tang, Peijie Dong, Zhenheng Tang, Xiaowen Chu, Junwei Liang

Combining CNNs or ViTs, with RNNs for spatiotemporal forecasting, has yielded unparalleled results in predicting temporal and spatial dynamics. However, modeling extensive global information remains a formidable challenge; CNNs are limited by their narrow receptive fields, and ViTs struggle with the intensive computational demands of their attention mechanisms. The emergence of recent Mamba-based architectures has been met with enthusiasm for their exceptional long-sequence modeling capabilities, surpassing established vision models in efficiency and accuracy, which motivates us to develop an innovative architecture tailored for spatiotemporal forecasting. In this paper, we propose the VMRNN cell, a new recurrent unit that integrates the strengths of Vision Mamba blocks with LSTM. We construct a network centered on VMRNN cells to tackle spatiotemporal prediction tasks effectively. Our extensive evaluations show that our proposed approach secures competitive results on a variety of tasks while maintaining a smaller model size. Our code is available at https://github.com/yyyujintang/VMRNN-PyTorch.

comment: 11 pages, 7 figures, 6 tables

♻ ☆ Image Captioning in news report scenario

Tianrui Liu, Qi Cai, Changxin Xu, Bo Hong, Jize Xiong, Yuxin Qiao, Tsungwei Yang

Image captioning strives to generate pertinent captions for specified images, situating itself at the crossroads of Computer Vision (CV) and Natural Language Processing (NLP). This endeavor is of paramount importance with far-reaching applications in recommendation systems, news outlets, social media, and beyond. Particularly within the realm of news reporting, captions are expected to encompass detailed information, such as the identities of celebrities captured in the images. However, much of the existing body of work primarily centers around understanding scenes and actions. In this paper, we explore the realm of image captioning specifically tailored for celebrity photographs, illustrating its broad potential for enhancing news industry practices. This exploration aims to augment automated news content generation, thereby facilitating a more nuanced dissemination of information. Our endeavor shows a broader horizon, enriching the narrative in news reporting through a more intuitive image captioning framework.

comment: 10 pages, 4 figures

♻ ☆ Gaze-guided Hand-Object Interaction Synthesis: Benchmark and Method

Jie Tian, Lingxiao Yang, Ran Ji, Yuexin Ma, Lan Xu, Jingyi Yu, Ye Shi, Jingya Wang

Gaze plays a crucial role in revealing human attention and intention, shedding light on the cognitive processes behind human actions. The integration of gaze guidance with the dynamics of hand-object interactions boosts the accuracy of human motion prediction. However, the lack of datasets that capture the intricate relationship and consistency among gaze, hand, and object movements remains a substantial hurdle. In this paper, we introduce the first Gaze-guided Hand-Object Interaction dataset, GazeHOI, and present a novel task for synthesizing gaze-guided hand-object interactions. Our dataset, GazeHOI, features simultaneous 3D modeling of gaze, hand, and object interactions, comprising 479 sequences with an average duration of 19.1 seconds, 812 sub-sequences, and 33 objects of various sizes. We propose a hierarchical framework centered on a gaze-guided hand-object interaction diffusion model, named GHO-Diffusion. In the pre-diffusion phase, we separate gaze conditions into spatial-temporal features and goal pose conditions at different levels of information granularity. During the diffusion phase, two gaze-conditioned diffusion models are stacked to simplify the complex synthesis of hand-object motions. Here, the object motion diffusion model generates sequences of object motions based on gaze conditions, while the hand motion diffusion model produces hand motions based on the generated object motion. To improve fine-grained goal pose alignment, we introduce a Spherical Gaussian constraint to guide the denoising step. In the subsequent post-diffusion phase, we optimize the generated hand motions using contact consistency. Our extensive experiments highlight the uniqueness of our dataset and the effectiveness of our approach.

♻ ☆ PKU-DyMVHumans: A Multi-View Video Benchmark for High-Fidelity Dynamic Human Modeling

Xiaoyun Zheng, Liwei Liao, Xufeng Li, Jianbo Jiao, Rongjie Wang, Feng Gao, Shiqi Wang, Ronggang Wang

High-quality human reconstruction and photo-realistic rendering of a dynamic scene is a long-standing problem in computer vision and graphics. Despite considerable efforts invested in developing various capture systems and reconstruction algorithms, recent advancements still struggle with loose or oversized clothing and overly complex poses. In part, this is due to the challenges of acquiring high-quality human datasets. To facilitate the development of these fields, in this paper, we present PKU-DyMVHumans, a versatile human-centric dataset for high-fidelity reconstruction and rendering of dynamic human scenarios from dense multi-view videos. It comprises 8.2 million frames captured by more than 56 synchronized cameras across diverse scenarios. These sequences comprise 32 human subjects across 45 different scenarios, each with a high-detailed appearance and realistic human motion. Inspired by recent advancements in neural radiance field (NeRF)-based scene representations, we carefully set up an off-the-shelf framework that is easy to provide those state-of-the-art NeRF-based implementations and benchmark on PKU-DyMVHumans dataset. It is paving the way for various applications like fine-grained foreground/background decomposition, high-quality human reconstruction and photo-realistic novel view synthesis of a dynamic scene. Extensive studies are performed on the benchmark, demonstrating new observations and challenges that emerge from using such high-fidelity dynamic data. The dataset is available at: https://pku-dymvhumans.github.io.

♻ ☆ X-Portrait: Expressive Portrait Animation with Hierarchical Motion Attention

You Xie, Hongyi Xu, Guoxian Song, Chao Wang, Yichun Shi, Linjie Luo

We propose X-Portrait, an innovative conditional diffusion model tailored for generating expressive and temporally coherent portrait animation. Specifically, given a single portrait as appearance reference, we aim to animate it with motion derived from a driving video, capturing both highly dynamic and subtle facial expressions along with wide-range head movements. As its core, we leverage the generative prior of a pre-trained diffusion model as the rendering backbone, while achieve fine-grained head pose and expression control with novel controlling signals within the framework of ControlNet. In contrast to conventional coarse explicit controls such as facial landmarks, our motion control module is learned to interpret the dynamics directly from the original driving RGB inputs. The motion accuracy is further enhanced with a patch-based local control module that effectively enhance the motion attention to small-scale nuances like eyeball positions. Notably, to mitigate the identity leakage from the driving signals, we train our motion control modules with scaling-augmented cross-identity images, ensuring maximized disentanglement from the appearance reference modules. Experimental results demonstrate the universal effectiveness of X-Portrait across a diverse range of facial portraits and expressive driving sequences, and showcase its proficiency in generating captivating portrait animations with consistently maintained identity characteristics.

♻ ☆ DiVa-360: The Dynamic Visual Dataset for Immersive Neural Fields

Cheng-You Lu, Peisen Zhou, Angela Xing, Chandradeep Pokhariya, Arnab Dey, Ishaan Shah, Rugved Mavidipalli, Dylan Hu, Andrew Comport, Kefan Chen, Srinath Sridhar

Advances in neural fields are enabling high-fidelity capture of the shape and appearance of dynamic 3D scenes. However, their capabilities lag behind those offered by conventional representations such as 2D videos because of algorithmic challenges and the lack of large-scale multi-view real-world datasets. We address the dataset limitation with DiVa-360, a real-world 360 dynamic visual dataset that contains synchronized high-resolution and long-duration multi-view video sequences of table-scale scenes captured using a customized low-cost system with 53 cameras. It contains 21 object-centric sequences categorized by different motion types, 25 intricate hand-object interaction sequences, and 8 long-duration sequences for a total of 17.4 M image frames. In addition, we provide foreground-background segmentation masks, synchronized audio, and text descriptions. We benchmark the state-of-the-art dynamic neural field methods on DiVa-360 and provide insights about existing methods and future challenges on long-duration neural field capture.

♻ ☆ HoloVIC: Large-scale Dataset and Benchmark for Multi-Sensor Holographic Intersection and Vehicle-Infrastructure Cooperative CVPR 2024

Cong Ma, Lei Qiao, Chengkai Zhu, Kai Liu, Zelong Kong, Qing Li, Xueqi Zhou, Yuheng Kan, Wei Wu

Vehicle-to-everything (V2X) is a popular topic in the field of Autonomous Driving in recent years. Vehicle-infrastructure cooperation (VIC) becomes one of the important research area. Due to the complexity of traffic conditions such as blind spots and occlusion, it greatly limits the perception capabilities of single-view roadside sensing systems. To further enhance the accuracy of roadside perception and provide better information to the vehicle side, in this paper, we constructed holographic intersections with various layouts to build a large-scale multi-sensor holographic vehicle-infrastructure cooperation dataset, called HoloVIC. Our dataset includes 3 different types of sensors (Camera, Lidar, Fisheye) and employs 4 sensor-layouts based on the different intersections. Each intersection is equipped with 6-18 sensors to capture synchronous data. While autonomous vehicles pass through these intersections for collecting VIC data. HoloVIC contains in total on 100k+ synchronous frames from different sensors. Additionally, we annotated 3D bounding boxes based on Camera, Fisheye, and Lidar. We also associate the IDs of the same objects across different devices and consecutive frames in sequence. Based on HoloVIC, we formulated four tasks to facilitate the development of related research. We also provide benchmarks for these tasks.

comment: Accept to CVPR 2024, Benchmark Website: https://holovic.net

♻ ☆ TRIPS: Trilinear Point Splatting for Real-Time Radiance Field Rendering

Linus Franke, Darius Rückert, Laura Fink, Marc Stamminger

Point-based radiance field rendering has demonstrated impressive results for novel view synthesis, offering a compelling blend of rendering quality and computational efficiency. However, also latest approaches in this domain are not without their shortcomings. 3D Gaussian Splatting [Kerbl and Kopanas et al. 2023] struggles when tasked with rendering highly detailed scenes, due to blurring and cloudy artifacts. On the other hand, ADOP [R\"uckert et al. 2022] can accommodate crisper images, but the neural reconstruction network decreases performance, it grapples with temporal instability and it is unable to effectively address large gaps in the point cloud. In this paper, we present TRIPS (Trilinear Point Splatting), an approach that combines ideas from both Gaussian Splatting and ADOP. The fundamental concept behind our novel technique involves rasterizing points into a screen-space image pyramid, with the selection of the pyramid layer determined by the projected point size. This approach allows rendering arbitrarily large points using a single trilinear write. A lightweight neural network is then used to reconstruct a hole-free image including detail beyond splat resolution. Importantly, our render pipeline is entirely differentiable, allowing for automatic optimization of both point sizes and positions. Our evaluation demonstrate that TRIPS surpasses existing state-of-the-art methods in terms of rendering quality while maintaining a real-time frame rate of 60 frames per second on readily available hardware. This performance extends to challenging scenarios, such as scenes featuring intricate geometry, expansive landscapes, and auto-exposed footage. The project page is located at: https://lfranke.github.io/trips/

♻ ☆ Semi-Supervised Crowd Counting from Unlabeled Data

Haoran Duan, Fan Wan, Rui Sun, Zeyu Wang, Varun Ojha, Yu Guan, Hubert P. H. Shum, Bingzhang Hu, Yang Long

Automatic Crowd behavior analysis can be applied to effectively help the daily transportation statistics and planning, which helps the smart city construction. As one of the most important keys, crowd counting has drawn increasing attention. Recent works achieved promising performance but relied on the supervised paradigm with expensive crowd annotations. To alleviate the annotation cost in real-world transportation scenarios, in this work we proposed a semi-supervised learning framework $S^{4}\textit{Crowd}$, which can leverage both unlabeled/labeled data for robust crowd counting. In the unsupervised pathway, two \textit{self-supervised losses} were proposed to simulate the crowd variations such as scale, illumination, based on which supervised information pseudo labels were generated and gradually refined. We also proposed a crowd-driven recurrent unit \textit{Gated-Crowd-Recurrent-Unit (GCRU)}, which can preserve discriminant crowd information by extracting second-order statistics, yielding pseudo labels with improved quality. A joint loss including both unsupervised/supervised information was proposed, and a dynamic weighting strategy was employed to balance the importance of the unsupervised loss and supervised loss at different training stages. We conducted extensive experiments on four popular crowd counting datasets in semi-supervised settings. Experimental results supported the effectiveness of each proposed component in our $S^{4}$Crowd framework. Our method achieved competitive performance in semi-supervised learning approaches on these crowd counting datasets.

♻ ☆ Efficient Pre-training for Localized Instruction Generation of Videos

Anil Batra, Davide Moltisanti, Laura Sevilla-Lara, Marcus Rohrbach, Frank Keller

Procedural videos show step-by-step demonstrations of tasks like recipe preparation. Understanding such videos is challenging, involving the precise localization of steps and the generation of textual instructions. Manually annotating steps and writing instructions is costly, which limits the size of current datasets and hinders effective learning. Leveraging large but noisy video-transcript datasets for pre-training can boost performance, but demands significant computational resources. Furthermore, transcripts contain irrelevant content and exhibit style variation compared to instructions written by human annotators. To mitigate both issues, we propose a technique, Sieve-&-Swap, to automatically curate a smaller dataset: (i) Sieve filters irrelevant transcripts and (ii) Swap enhances the quality of the text instruction by automatically replacing the transcripts with human-written instructions from a text-only recipe dataset. The curated dataset, three orders of magnitude smaller than current web-scale datasets, enables efficient training of large-scale models with competitive performance. We complement our Sieve-\&-Swap approach with a Procedure Transformer (ProcX) for end-to-end step localization and instruction generation for procedural videos. When this model is pre-trained on our curated dataset, it achieves state-of-the-art performance in zero-shot and finetuning settings on YouCook2 and Tasty, while using a fraction of the computational resources.

comment: This version has some missing experiments and elaborative technical details

♻ ☆ SimLVSeg: Simplifying Left Ventricular Segmentation in 2D+Time Echocardiograms with Self- and Weakly-Supervised Learning

Fadillah Maani, Asim Ukaye, Nada Saadi, Numan Saeed, Mohammad Yaqub

Echocardiography has become an indispensable clinical imaging modality for general heart health assessment. From calculating biomarkers such as ejection fraction to the probability of a patient's heart failure, accurate segmentation of the heart structures allows doctors to assess the heart's condition and devise treatments with greater precision and accuracy. However, achieving accurate and reliable left ventricle segmentation is time-consuming and challenging due to different reasons. Hence, clinicians often rely on segmenting the left ventricular (LV) in two specific echocardiogram frames to make a diagnosis. This limited coverage in manual LV segmentation poses a challenge for developing automatic LV segmentation with high temporal consistency, as the resulting dataset is typically annotated sparsely. In response to this challenge, this work introduces SimLVSeg, a novel paradigm that enables video-based networks for consistent LV segmentation from sparsely annotated echocardiogram videos. SimLVSeg consists of self-supervised pre-training with temporal masking, followed by weakly supervised learning tailored for LV segmentation from sparse annotations. We demonstrate how SimLVSeg outperforms the state-of-the-art solutions by achieving a 93.32% (95%CI 93.21-93.43%) dice score on the largest 2D+time echocardiography dataset (EchoNet-Dynamic) while being more efficient. SimLVSeg is compatible with two types of video segmentation networks: 2D super image and 3D segmentation. To show the effectiveness of our approach, we provide extensive ablation studies, including pre-training settings and various deep learning backbones. We further conduct an out-of-distribution test to showcase SimLVSeg's generalizability on unseen distribution (CAMUS dataset). The code is publicly available at https://github.com/fadamsyah/SimLVSeg.

♻ ☆ HIMap: HybrId Representation Learning for End-to-end Vectorized HD Map Construction CVPR 2024

Yi Zhou, Hui Zhang, Jiaqian Yu, Yifan Yang, Sangil Jung, Seung-In Park, ByungIn Yoo

Vectorized High-Definition (HD) map construction requires predictions of the category and point coordinates of map elements (e.g. road boundary, lane divider, pedestrian crossing, etc.). State-of-the-art methods are mainly based on point-level representation learning for regressing accurate point coordinates. However, this pipeline has limitations in obtaining element-level information and handling element-level failures, e.g. erroneous element shape or entanglement between elements. To tackle the above issues, we propose a simple yet effective HybrId framework named HIMap to sufficiently learn and interact both point-level and element-level information. Concretely, we introduce a hybrid representation called HIQuery to represent all map elements, and propose a point-element interactor to interactively extract and encode the hybrid information of elements, e.g. point position and element shape, into the HIQuery. Additionally, we present a point-element consistency constraint to enhance the consistency between the point-level and element-level information. Finally, the output point-element integrated HIQuery can be directly converted into map elements' class, point coordinates, and mask. We conduct extensive experiments and consistently outperform previous methods on both nuScenes and Argoverse2 datasets. Notably, our method achieves $77.8$ mAP on the nuScenes dataset, remarkably superior to previous SOTAs by $8.3$ mAP at least.

comment: Accepted to CVPR 2024

♻ ☆ Exploiting Semantic Reconstruction to Mitigate Hallucinations in Vision-Language Models

Minchan Kim, Minyeong Kim, Junik Bae, Suhwan Choi, Sungkyung Kim, Buru Chang

Hallucinations in vision-language models pose a significant challenge to their reliability, particularly in the generation of long captions. Current methods fall short of accurately identifying and mitigating these hallucinations. To address this issue, we introduce ESREAL, a novel unsupervised learning framework designed to suppress the generation of hallucinations through accurate localization and penalization of hallucinated tokens. Initially, ESREAL creates a reconstructed image based on the generated caption and aligns its corresponding regions with those of the original image. This semantic reconstruction aids in identifying both the presence and type of token-level hallucinations within the generated caption. Subsequently, ESREAL computes token-level hallucination scores by assessing the semantic similarity of aligned regions based on the type of hallucination. Finally, ESREAL employs a proximal policy optimization algorithm, where it selectively penalizes hallucinated tokens according to their token-level hallucination scores. Our framework notably reduces hallucinations in LLaVA, InstructBLIP, and mPLUG-Owl2 by 32.81%, 27.08%, and 7.46% on the CHAIR metric. This improvement is achieved solely through signals derived from the image itself, without the need for any image-text pairs.

♻ ☆ Pushing Auto-regressive Models for 3D Shape Generation at Capacity and Scalability

Xuelin Qian, Yu Wang, Simian Luo, Yinda Zhang, Ying Tai, Zhenyu Zhang, Chengjie Wang, Xiangyang Xue, Bo Zhao, Tiejun Huang, Yunsheng Wu, Yanwei Fu

Auto-regressive models have achieved impressive results in 2D image generation by modeling joint distributions in grid space. In this paper, we extend auto-regressive models to 3D domains, and seek a stronger ability of 3D shape generation by improving auto-regressive models at capacity and scalability simultaneously. Firstly, we leverage an ensemble of publicly available 3D datasets to facilitate the training of large-scale models. It consists of a comprehensive collection of approximately 900,000 objects, with multiple properties of meshes, points, voxels, rendered images, and text captions. This diverse labeled dataset, termed Objaverse-Mix, empowers our model to learn from a wide range of object variations. However, directly applying 3D auto-regression encounters critical challenges of high computational demands on volumetric grids and ambiguous auto-regressive order along grid dimensions, resulting in inferior quality of 3D shapes. To this end, we then present a novel framework Argus3D in terms of capacity. Concretely, our approach introduces discrete representation learning based on a latent vector instead of volumetric grids, which not only reduces computational costs but also preserves essential geometric details by learning the joint distributions in a more tractable order. The capacity of conditional generation can thus be realized by simply concatenating various conditioning inputs to the latent vector, such as point clouds, categories, images, and texts. In addition, thanks to the simplicity of our model architecture, we naturally scale up our approach to a larger model with an impressive 3.6 billion parameters, further enhancing the quality of versatile 3D generation. Extensive experiments on four generation tasks demonstrate that Argus3D can synthesize diverse and faithful shapes across multiple categories, achieving remarkable performance.

comment: Project page: https://argus-3d.github.io/ . Datasets: https://huggingface.co/datasets/BAAI/Objaverse-MIX. arXiv admin note: substantial text overlap with arXiv:2303.14700

♻ ☆ ReMoS: 3D Motion-Conditioned Reaction Synthesis for Two-Person Interactions

Anindita Ghosh, Rishabh Dabral, Vladislav Golyanik, Christian Theobalt, Philipp Slusallek

Current approaches for 3D human motion synthesis generate high-quality animations of digital humans performing a wide variety of actions and gestures. However, a notable technological gap exists in addressing the complex dynamics of multi-human interactions within this paradigm. In this work, we present ReMoS, a denoising diffusion-based model that synthesizes full-body reactive motion of a person in a two-person interaction scenario. Assuming the motion of one person is given, we employ a combined spatio-temporal cross-attention mechanism to synthesize the reactive body and hand motion of the second person, thereby completing the interactions between the two. We demonstrate ReMoS across challenging two-person scenarios such as pair-dancing, Ninjutsu, kickboxing, and acrobatics, where one person's movements have complex and diverse influences on the other. We also contribute the ReMoCap dataset for two-person interactions containing full-body and finger motions. We evaluate ReMoS through multiple quantitative metrics, qualitative visualizations, and a user study, and also indicate usability in interactive motion editing applications.

comment: 17 pages, 7 figures, 5 tables

♻ ☆ MedPromptX: Grounded Multimodal Prompting for Chest X-ray Diagnosis

Mai A. Shaaban, Adnan Khan, Mohammad Yaqub

Chest X-ray images are commonly used for predicting acute and chronic cardiopulmonary conditions, but efforts to integrate them with structured clinical data face challenges due to incomplete electronic health records (EHR). This paper introduces \textbf{MedPromptX}, the first model to integrate multimodal large language models (MLLMs), few-shot prompting (FP) and visual grounding (VG) to combine imagery with EHR data for chest X-ray diagnosis. A pre-trained MLLM is utilized to complement the missing EHR information, providing a comprehensive understanding of patients' medical history. Additionally, FP reduces the necessity for extensive training of MLLMs while effectively tackling the issue of hallucination. Nevertheless, the process of determining the optimal number of few-shot examples and selecting high-quality candidates can be burdensome, yet it profoundly influences model performance. Hence, we propose a new technique that dynamically refines few-shot data for real-time adjustment to new patient scenarios. Moreover, VG aids in focusing the model's attention on relevant regions of interest in X-ray images, enhancing the identification of abnormalities. We release MedPromptX-VQA, a new in-context visual question answering dataset encompassing interleaved image and EHR data derived from MIMIC-IV and MIMIC-CXR databases. Results demonstrate the SOTA performance of MedPromptX, achieving an 11% improvement in F1-score compared to the baselines. Code and data are available at https://github.com/BioMedIA-MBZUAI/MedPromptX

♻ ☆ Text-Guided Variational Image Generation for Industrial Anomaly Detection and Segmentation CVPR 2024

Mingyu Lee, Jongwon Choi

We propose a text-guided variational image generation method to address the challenge of getting clean data for anomaly detection in industrial manufacturing. Our method utilizes text information about the target object, learned from extensive text library documents, to generate non-defective data images resembling the input image. The proposed framework ensures that the generated non-defective images align with anticipated distributions derived from textual and image-based knowledge, ensuring stability and generality. Experimental results demonstrate the effectiveness of our approach, surpassing previous methods even with limited non-defective data. Our approach is validated through generalization tests across four baseline models and three distinct datasets. We present an additional analysis to enhance the effectiveness of anomaly detection models by utilizing the generated images.

comment: 18 pages, Accepted to CVPR 2024

♻ ☆ Identity-aware Dual-constraint Network for Cloth-Changing Person Re-identification

Peini Guo, Mengyuan Liu, Hong Liu, Ruijia Fan, Guoquan Wang, Bin He

Cloth-Changing Person Re-Identification (CC-ReID) aims to accurately identify the target person in more realistic surveillance scenarios, where pedestrians usually change their clothing. Despite great progress, limited cloth-changing training samples in existing CC-ReID datasets still prevent the model from adequately learning cloth-irrelevant features. In addition, due to the absence of explicit supervision to keep the model constantly focused on cloth-irrelevant areas, existing methods are still hampered by the disruption of clothing variations. To solve the above issues, we propose an Identity-aware Dual-constraint Network (IDNet) for the CC-ReID task. Specifically, to help the model extract cloth-irrelevant clues, we propose a Clothes Diversity Augmentation (CDA), which generates more realistic cloth-changing samples by enriching the clothing color while preserving the texture. In addition, a Multi-scale Constraint Block (MCB) is designed, which extracts fine-grained identity-related features and effectively transfers cloth-irrelevant knowledge. Moreover, a Counterfactual-guided Attention Module (CAM) is presented, which learns cloth-irrelevant features from channel and space dimensions and utilizes the counterfactual intervention for supervising the attention map to highlight identity-related regions. Finally, a Semantic Alignment Constraint (SAC) is designed to facilitate high-level semantic feature interaction. Comprehensive experiments on four CC-ReID datasets indicate that our method outperforms prior state-of-the-art approaches.

♻ ☆ Unveiling the Pitfalls of Knowledge Editing for Large Language Models ICLR 2024

Zhoubo Li, Ningyu Zhang, Yunzhi Yao, Mengru Wang, Xi Chen, Huajun Chen

As the cost associated with fine-tuning Large Language Models (LLMs) continues to rise, recent research efforts have pivoted towards developing methodologies to edit implicit knowledge embedded within LLMs. Yet, there's still a dark cloud lingering overhead -- will knowledge editing trigger butterfly effect? since it is still unclear whether knowledge editing might introduce side effects that pose potential risks or not. This paper pioneers the investigation into the potential pitfalls associated with knowledge editing for LLMs. To achieve this, we introduce new benchmark datasets and propose innovative evaluation metrics. Our results underline two pivotal concerns: (1) Knowledge Conflict: Editing groups of facts that logically clash can magnify the inherent inconsistencies in LLMs-a facet neglected by previous methods. (2) Knowledge Distortion: Altering parameters with the aim of editing factual knowledge can irrevocably warp the innate knowledge structure of LLMs. Experimental results vividly demonstrate that knowledge editing might inadvertently cast a shadow of unintended consequences on LLMs, which warrant attention and efforts for future works. Code and data are available at https://github.com/zjunlp/PitfallsKnowledgeEditing.

comment: ICLR 2024

♻ ☆ InNeRF360: Text-Guided 3D-Consistent Object Inpainting on 360-degree Neural Radiance Fields CVPR 2024

Dongqing Wang, Tong Zhang, Alaa Abboud, Sabine Süsstrunk

We propose InNeRF360, an automatic system that accurately removes text-specified objects from 360-degree Neural Radiance Fields (NeRF). The challenge is to effectively remove objects while inpainting perceptually consistent content for the missing regions, which is particularly demanding for existing NeRF models due to their implicit volumetric representation. Moreover, unbounded scenes are more prone to floater artifacts in the inpainted region than frontal-facing scenes, as the change of object appearance and background across views is more sensitive to inaccurate segmentations and inconsistent inpainting. With a trained NeRF and a text description, our method efficiently removes specified objects and inpaints visually consistent content without artifacts. We apply depth-space warping to enforce consistency across multiview text-encoded segmentations, and then refine the inpainted NeRF model using perceptual priors and 3D diffusion-based geometric priors to ensure visual plausibility. Through extensive experiments in segmentation and inpainting on 360-degree and frontal-facing NeRFs, we show that our approach is effective and enhances NeRF's editability. Project page: https://ivrl.github.io/InNeRF360.

comment: CVPR 2024

♻ ☆ Passive Non-Line-of-Sight Imaging with Light Transport Modulation

Jiarui Zhang, Ruixu Geng, Xiaolong Du, Yan Chen, Houqiang Li, Yang Hu

Passive non-line-of-sight (NLOS) imaging has witnessed rapid development in recent years, due to its ability to image objects that are out of sight. The light transport condition plays an important role in this task since changing the conditions will lead to different imaging models. Existing learning-based NLOS methods usually train independent models for different light transport conditions, which is computationally inefficient and impairs the practicality of the models. In this work, we propose NLOS-LTM, a novel passive NLOS imaging method that effectively handles multiple light transport conditions with a single network. We achieve this by inferring a latent light transport representation from the projection image and using this representation to modulate the network that reconstructs the hidden image from the projection image. We train a light transport encoder together with a vector quantizer to obtain the light transport representation. To further regulate this representation, we jointly learn both the reconstruction network and the reprojection network during training. A set of light transport modulation blocks is used to modulate the two jointly trained networks in a multi-scale way. Extensive experiments on a large-scale passive NLOS dataset demonstrate the superiority of the proposed method. The code is available at https://github.com/JerryOctopus/NLOS-LTM.

♻ ★ ViT-Lens: Towards Omni-modal Representations CVPR2024

Weixian Lei, Yixiao Ge, Kun Yi, Jianfeng Zhang, Difei Gao, Dylan Sun, Yuying Ge, Ying Shan, Mike Zheng Shou

Aiming to advance AI agents, large foundation models significantly improve reasoning and instruction execution, yet the current focus on vision and language neglects the potential of perceiving diverse modalities in open-world environments. However, the success of data-driven vision and language models is costly or even infeasible to be reproduced for rare modalities. In this paper, we present ViT-Lens-2 that facilitates efficient omni-modal representation learning by perceiving novel modalities with a pretrained ViT and aligning them to a pre-defined space. Specifically, the modality-specific lens is tuned to project any-modal signals to an intermediate embedding space, which are then processed by a strong ViT with pre-trained visual knowledge. The encoded representations are optimized toward aligning with the modal-independent space, pre-defined by off-the-shelf foundation models. ViT-Lens-2 provides a unified solution for representation learning of increasing modalities with two appealing advantages: (i) Unlocking the great potential of pretrained ViTs to novel modalities effectively with efficient data regime; (ii) Enabling emergent downstream capabilities through modality alignment and shared ViT parameters. We tailor ViT-Lens-2 to learn representations for 3D point cloud, depth, audio, tactile and EEG, and set new state-of-the-art results across various understanding tasks, such as zero-shot classification. By seamlessly integrating ViT-Lens-2 into Multimodal Foundation Models, we enable Any-modality to Text and Image Generation in a zero-shot manner. Code and models are available at https://github.com/TencentARC/ViT-Lens.

comment: This work is a follow-up of arXiv:2308.10185. Accepted to CVPR2024

♻ ☆ Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification CVPR 2024

Kaijie Ren, Lei Zhang

Visible-Infrared Person Re-identification (VI-ReID) is a challenging cross-modal pedestrian retrieval task, due to significant intra-class variations and cross-modal discrepancies among different cameras. Existing works mainly focus on embedding images of different modalities into a unified space to mine modality-shared features. They only seek distinctive information within these shared features, while ignoring the identity-aware useful information that is implicit in the modality-specific features. To address this issue, we propose a novel Implicit Discriminative Knowledge Learning (IDKL) network to uncover and leverage the implicit discriminative information contained within the modality-specific. First, we extract modality-specific and modality-shared features using a novel dual-stream network. Then, the modality-specific features undergo purification to reduce their modality style discrepancies while preserving identity-aware discriminative knowledge. Subsequently, this kind of implicit knowledge is distilled into the modality-shared feature to enhance its distinctiveness. Finally, an alignment loss is proposed to minimize modality discrepancy on enhanced modality-shared features. Extensive experiments on multiple public datasets demonstrate the superiority of IDKL network over the state-of-the-art methods. Code is available at https://github.com/1KK077/IDKL.

comment: CVPR 2024

♻ ☆ In Search of a Data Transformation That Accelerates Neural Field Training CVPR 2024

Junwon Seo, Sangyoon Lee, Kwang In Kim, Jaeho Lee

Neural field is an emerging paradigm in data representation that trains a neural network to approximate the given signal. A key obstacle that prevents its widespread adoption is the encoding speed-generating neural fields requires an overfitting of a neural network, which can take a significant number of SGD steps to reach the desired fidelity level. In this paper, we delve into the impacts of data transformations on the speed of neural field training, specifically focusing on how permuting pixel locations affect the convergence speed of SGD. Counterintuitively, we find that randomly permuting the pixel locations can considerably accelerate the training. To explain this phenomenon, we examine the neural field training through the lens of PSNR curves, loss landscapes, and error patterns. Our analyses suggest that the random pixel permutations remove the easy-to-fit patterns, which facilitate easy optimization in the early stage but hinder capturing fine details of the signal.

comment: CVPR 2024

♻ ☆ AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation CVPR 2024

Jeongsoo Choi, Se Jin Park, Minsu Kim, Yong Man Ro

This paper proposes a novel direct Audio-Visual Speech to Audio-Visual Speech Translation (AV2AV) framework, where the input and output of the system are multimodal (i.e., audio and visual speech). With the proposed AV2AV, two key advantages can be brought: 1) We can perform real-like conversations with individuals worldwide in a virtual meeting by utilizing our own primary languages. In contrast to Speech-to-Speech Translation (A2A), which solely translates between audio modalities, the proposed AV2AV directly translates between audio-visual speech. This capability enhances the dialogue experience by presenting synchronized lip movements along with the translated speech. 2) We can improve the robustness of the spoken language translation system. By employing the complementary information of audio-visual speech, the system can effectively translate spoken language even in the presence of acoustic noise, showcasing robust performance. To mitigate the problem of the absence of a parallel AV2AV translation dataset, we propose to train our spoken language translation system with the audio-only dataset of A2A. This is done by learning unified audio-visual speech representations through self-supervised learning in advance to train the translation system. Moreover, we propose an AV-Renderer that can generate raw audio and video in parallel. It is designed with zero-shot speaker modeling, thus the speaker in source audio-visual speech can be maintained at the target translated audio-visual speech. The effectiveness of AV2AV is evaluated with extensive experiments in a many-to-many language translation setting. Demo page is available on https://choijeongsoo.github.io/av2av.

comment: CVPR 2024. Code & Demo: https://choijeongsoo.github.io/av2av

♻ ☆ SINC: Spatial Composition of 3D Human Motions for Simultaneous Action Generation

Nikos Athanasiou, Mathis Petrovich, Michael J. Black, Gül Varol

Our goal is to synthesize 3D human motions given textual inputs describing simultaneous actions, for example 'waving hand' while 'walking' at the same time. We refer to generating such simultaneous movements as performing 'spatial compositions'. In contrast to temporal compositions that seek to transition from one action to another, spatial compositing requires understanding which body parts are involved in which action, to be able to move them simultaneously. Motivated by the observation that the correspondence between actions and body parts is encoded in powerful language models, we extract this knowledge by prompting GPT-3 with text such as "what are the body parts involved in the action ?", while also providing the parts list and few-shot examples. Given this action-part mapping, we combine body parts from two motions together and establish the first automated method to spatially compose two actions. However, training data with compositional actions is always limited by the combinatorics. Hence, we further create synthetic data with this approach, and use it to train a new state-of-the-art text-to-motion generation model, called SINC ("SImultaneous actioN Compositions for 3D human motions"). In our experiments, that training with such GPT-guided synthetic data improves spatial composition generation over baselines. Our code is publicly available at https://sinc.is.tue.mpg.de/.

comment: Teaser Fixed

♻ ☆ Powerful Lossy Compression for Noisy Images ICME 2024

Shilv Cai, Xiaoguo Liang, Shuning Cao, Luxin Yan, Sheng Zhong, Liqun Chen, Xu Zou

Image compression and denoising represent fundamental challenges in image processing with many real-world applications. To address practical demands, current solutions can be categorized into two main strategies: 1) sequential method; and 2) joint method. However, sequential methods have the disadvantage of error accumulation as there is information loss between multiple individual models. Recently, the academic community began to make some attempts to tackle this problem through end-to-end joint methods. Most of them ignore that different regions of noisy images have different characteristics. To solve these problems, in this paper, our proposed signal-to-noise ratio~(SNR) aware joint solution exploits local and non-local features for image compression and denoising simultaneously. We design an end-to-end trainable network, which includes the main encoder branch, the guidance branch, and the signal-to-noise ratio~(SNR) aware branch. We conducted extensive experiments on both synthetic and real-world datasets, demonstrating that our joint solution outperforms existing state-of-the-art methods.

comment: Accepted by ICME 2024

♻ ☆ ViT-Lens: Initiating Omni-Modal Exploration through 3D Insights

Weixian Lei, Yixiao Ge, Jianfeng Zhang, Dylan Sun, Kun Yi, Ying Shan, Mike Zheng Shou

Though the success of CLIP-based training recipes in vision-language models, their scalability to more modalities (e.g., 3D, audio, etc.) is limited to large-scale data, which is expensive or even inapplicable for rare modalities. In this paper, we present ViT-Lens that facilitates efficient omni-modal representation learning by perceiving novel modalities with a pretrained ViT and aligning to a pre-defined space. Specifically, the modality-specific lens is tuned to project multimodal signals to the shared embedding space, which are then processed by a strong ViT that carries pre-trained image knowledge. The encoded multimodal representations are optimized toward aligning with the modal-independent space, pre-defined by off-the-shelf foundation models. A well-trained lens with a ViT backbone has the potential to serve as one of these foundation models, supervising the learning of subsequent modalities. ViT-Lens provides a unified solution for representation learning of increasing modalities with two appealing benefits: (i) Exploiting the pretrained ViT across tasks and domains effectively with efficient data regime; (ii) Emergent downstream capabilities of novel modalities are demonstrated due to the modality alignment space. We evaluate ViT-Lens in the context of 3D as an initial verification. In zero-shot 3D classification, ViT-Lens achieves substantial improvements over previous state-of-the-art, showing 52.0% accuracy on Objaverse-LVIS, 87.4% on ModelNet40, and 60.6% on ScanObjectNN. Furthermore, we enable zero-shot 3D question-answering by simply integrating the trained 3D lens into the InstructBLIP model without any adaptation. We will release the results of ViT-Lens on more modalities in the near future.

comment: 19 pages, 4 figures and 9 tables

♻ ☆ TP2O: Creative Text Pair-to-Object Generation using Balance Swap-Sampling

Jun Li, Zedong Zhang, Jian Yang

Generating creative combinatorial objects from two seemingly unrelated object texts is a challenging task in text-to-image synthesis, often hindered by a focus on emulating existing data distributions. In this paper, we develop a straightforward yet highly effective method, called \textbf{balance swap-sampling}. First, we propose a swapping mechanism that generates a novel combinatorial object image set by randomly exchanging intrinsic elements of two text embeddings through a cutting-edge diffusion model. Second, we introduce a balance swapping region to efficiently sample a small subset from the newly generated image set by balancing CLIP distances between the new images and their original generations, increasing the likelihood of accepting the high-quality combinations. Last, we employ a segmentation method to compare CLIP distances among the segmented components, ultimately selecting the most promising object from the sampled subset. Extensive experiments demonstrate that our approach outperforms recent SOTA T2I methods. Surprisingly, our results even rival those of human artists, such as frog-broccoli.

comment: Project page: https://tp2o.github.io/anon/

♻ ☆ Segment and Caption Anything CVPR 24

Xiaoke Huang, Jianfeng Wang, Yansong Tang, Zheng Zhang, Han Hu, Jiwen Lu, Lijuan Wang, Zicheng Liu

We propose a method to efficiently equip the Segment Anything Model (SAM) with the ability to generate regional captions. SAM presents strong generalizability to segment anything while is short for semantic understanding. By introducing a lightweight query-based feature mixer, we align the region-specific features with the embedding space of language models for later caption generation. As the number of trainable parameters is small (typically in the order of tens of millions), it costs less computation, less memory usage, and less communication bandwidth, resulting in both fast and scalable training. To address the scarcity problem of regional caption data, we propose to first pre-train our model on objection detection and segmentation tasks. We call this step weak supervision pretraining since the pre-training data only contains category names instead of full-sentence descriptions. The weak supervision pretraining allows us to leverage many publicly available object detection and segmentation datasets. We conduct extensive experiments to demonstrate the superiority of our method and validate each design choice. This work serves as a stepping stone towards scaling up regional captioning data and sheds light on exploring efficient ways to augment SAM with regional semantics. The project page, along with the associated code, can be accessed via https://xk-huang.github.io/segment-caption-anything/.

comment: The project page, along with the associated code, can be accessed via https://xk-huang.github.io/segment-caption-anything/; Update author information; Accepted by CVPR 24

♻ ☆ TagAlign: Improving Vision-Language Alignment with Multi-Tag Classification

Qinying Liu, Wei Wu, Kecheng Zheng, Zhan Tong, Jiawei Liu, Yu Liu, Wei Chen, Zilei Wang, Yujun Shen

The crux of learning vision-language models is to extract semantically aligned information from visual and linguistic data. Existing attempts usually face the problem of coarse alignment, e.g., the vision encoder struggles in localizing an attribute-specified object. In this work, we propose an embarrassingly simple approach to better align image and text features with no need of additional data formats other than image-text pairs. Concretely, given an image and its paired text, we manage to parse objects (e.g., cat) and attributes (e.g., black) from the description, which are highly likely to exist in the image. It is noteworthy that the parsing pipeline is fully automatic and thus enjoys good scalability. With these parsed semantics as supervision signals, we can complement the commonly used image-text contrastive loss with the multi-tag classification loss. Extensive experimental results on a broad suite of semantic segmentation datasets substantiate the average 5.2\% improvement of our framework over existing alternatives. Furthermore, the visualization results indicate that attribute supervision makes vision-language models accurately localize attribute-specified objects. Project page can be found at https://qinying-liu.github.io/Tag-Align.

♻ ☆ SGS-SLAM: Semantic Gaussian Splatting For Neural Dense SLAM

Mingrui Li, Shuhong Liu, Heng Zhou, Guohao Zhu, Na Cheng, Tianchen Deng, Hongyu Wang

We present SGS-SLAM, the first semantic visual SLAM system based on Gaussian Splatting. It incorporates appearance, geometry, and semantic features through multi-channel optimization, addressing the oversmoothing limitations of neural implicit SLAM systems in high-quality rendering, scene understanding, and object-level geometry. We introduce a unique semantic feature loss that effectively compensates for the shortcomings of traditional depth and color losses in object optimization. Through a semantic-guided keyframe selection strategy, we prevent erroneous reconstructions caused by cumulative errors. Extensive experiments demonstrate that SGS-SLAM delivers state-of-the-art performance in camera pose estimation, map reconstruction, precise semantic segmentation, and object-level geometric accuracy, while ensuring real-time rendering capabilities.

♻ ☆ ArtAdapter: Text-to-Image Style Transfer using Multi-Level Style Encoder and Explicit Adaptation

Dar-Yen Chen, Hamish Tennent, Ching-Wen Hsu

This work introduces ArtAdapter, a transformative text-to-image (T2I) style transfer framework that transcends traditional limitations of color, brushstrokes, and object shape, capturing high-level style elements such as composition and distinctive artistic expression. The integration of a multi-level style encoder with our proposed explicit adaptation mechanism enables ArtAdapter to achieve unprecedented fidelity in style transfer, ensuring close alignment with textual descriptions. Additionally, the incorporation of an Auxiliary Content Adapter (ACA) effectively separates content from style, alleviating the borrowing of content from style references. Moreover, our novel fast finetuning approach could further enhance zero-shot style representation while mitigating the risk of overfitting. Comprehensive evaluations confirm that ArtAdapter surpasses current state-of-the-art methods.

♻ ☆ Clean-image Backdoor Attacks

Dazhong Rong, Guoyao Yu, Shuheng Shen, Xinyi Fu, Peng Qian, Jianhai Chen, Qinming He, Xing Fu, Weiqiang Wang

To gather a significant quantity of annotated training data for high-performance image classification models, numerous companies opt to enlist third-party providers to label their unlabeled data. This practice is widely regarded as secure, even in cases where some annotated errors occur, as the impact of these minor inaccuracies on the final performance of the models is negligible and existing backdoor attacks require attacker's ability to poison the training images. Nevertheless, in this paper, we propose clean-image backdoor attacks which uncover that backdoors can still be injected via a fraction of incorrect labels without modifying the training images. Specifically, in our attacks, the attacker first seeks a trigger feature to divide the training images into two parts: those with the feature and those without it. Subsequently, the attacker falsifies the labels of the former part to a backdoor class. The backdoor will be finally implanted into the target model after it is trained on the poisoned data. During the inference phase, the attacker can activate the backdoor in two ways: slightly modifying the input image to obtain the trigger feature, or taking an image that naturally has the trigger feature as input. We conduct extensive experiments to demonstrate the effectiveness and practicality of our attacks. According to the experimental results, we conclude that our attacks seriously jeopardize the fairness and robustness of image classification models, and it is necessary to be vigilant about the incorrect labels in outsourced labeling.

♻ ☆ Transferring Relative Monocular Depth to Surgical Vision with Temporal Consistency

Charlie Budd, Tom Vercauteren

Relative monocular depth, inferring depth up to shift and scale from a single image, is an active research topic. Recent deep learning models, trained on large and varied meta-datasets, now provide excellent performance in the domain of natural images. However, few datasets exist which provide ground truth depth for endoscopic images, making training such models from scratch unfeasible. This work investigates the transfer of these models into the surgical domain, and presents an effective and simple way to improve on standard supervision through the use of temporal consistency self-supervision. We show temporal consistency significantly improves supervised training alone when transferring to the low-data regime of endoscopy, and outperforms the prevalent self-supervision technique for this task. In addition we show our method drastically outperforms the state-of-the-art method from within the domain of endoscopy. We also release our code, model and ensembled meta-dataset, Meta-MED, establishing a strong benchmark for future work.

♻ ☆ Towards Source-free Domain Adaptive Semantic Segmentation via Importance-aware and Prototype-contrast Learning

Yihong Cao, Hui Zhang, Xiao Lu, Zheng Xiao, Kailun Yang, Yaonan Wang

Domain adaptive semantic segmentation enables robust pixel-wise understanding in real-world driving scenes. Source-free domain adaptation, as a more practical technique, addresses the concerns of data privacy and storage limitations in typical unsupervised domain adaptation methods, making it especially relevant in the context of intelligent vehicles. It utilizes a well-trained source model and unlabeled target data to achieve adaptation in the target domain. However, in the absence of source data and target labels, current solutions cannot sufficiently reduce the impact of domain shift and fully leverage the information from the target data. In this paper, we propose an end-to-end source-free domain adaptation semantic segmentation method via Importance-Aware and Prototype-Contrast (IAPC) learning. The proposed IAPC framework effectively extracts domain-invariant knowledge from the well-trained source model and learns domain-specific knowledge from the unlabeled target domain. Specifically, considering the problem of domain shift in the prediction of the target domain by the source model, we put forward an importance-aware mechanism for the biased target prediction probability distribution to extract domain-invariant knowledge from the source model. We further introduce a prototype-contrast strategy, which includes a prototype-symmetric cross-entropy loss and a prototype-enhanced cross-entropy loss, to learn target intra-domain knowledge without relying on labels. A comprehensive variety of experiments on two domain adaptive semantic segmentation benchmarks demonstrates that the proposed end-to-end IAPC solution outperforms existing state-of-the-art methods. The source code is publicly available at https://github.com/yihong-97/Source-free-IAPC.

comment: Accepted to IEEE Transactions on Intelligent Vehicles (T-IV). The source code is publicly available at https://github.com/yihong-97/Source-free-IAPC

♻ ☆ SD4Match: Learning to Prompt Stable Diffusion Model for Semantic Matching CVPR 2024

Xinghui Li, Jingyi Lu, Kai Han, Victor Prisacariu

In this paper, we address the challenge of matching semantically similar keypoints across image pairs. Existing research indicates that the intermediate output of the UNet within the Stable Diffusion (SD) can serve as robust image feature maps for such a matching task. We demonstrate that by employing a basic prompt tuning technique, the inherent potential of Stable Diffusion can be harnessed, resulting in a significant enhancement in accuracy over previous approaches. We further introduce a novel conditional prompting module that conditions the prompt on the local details of the input image pairs, leading to a further improvement in performance. We designate our approach as SD4Match, short for Stable Diffusion for Semantic Matching. Comprehensive evaluations of SD4Match on the PF-Pascal, PF-Willow, and SPair-71k datasets show that it sets new benchmarks in accuracy across all these datasets. Particularly, SD4Match outperforms the previous state-of-the-art by a margin of 12 percentage points on the challenging SPair-71k dataset.

comment: Accepted to CVPR 2024. Project website: https://sd4match.active.vision/

♻ ☆ ObjectCompose: Evaluating Resilience of Vision-Based Models on Object-to-Background Compositional Changes

Hashmat Shadab Malik, Muhammad Huzaifa, Muzammal Naseer, Salman Khan, Fahad Shahbaz Khan

Given the large-scale multi-modal training of recent vision-based models and their generalization capabilities, understanding the extent of their robustness is critical for their real-world deployment. In this work, we evaluate the resilience of current vision-based models against diverse object-to-background context variations. The majority of robustness evaluation methods have introduced synthetic datasets to induce changes to object characteristics (viewpoints, scale, color) or utilized image transformation techniques (adversarial changes, common corruptions) on real images to simulate shifts in distributions. Recent works have explored leveraging large language models and diffusion models to generate changes in the background. However, these methods either lack in offering control over the changes to be made or distort the object semantics, making them unsuitable for the task. Our method, on the other hand, can induce diverse object-to-background changes while preserving the original semantics and appearance of the object. To achieve this goal, we harness the generative capabilities of text-to-image, image-to-text, and image-to-segment models to automatically generate a broad spectrum of object-to-background changes. We induce both natural and adversarial background changes by either modifying the textual prompts or optimizing the latents and textual embedding of text-to-image models. We produce various versions of standard vision datasets (ImageNet, COCO), incorporating either diverse and realistic backgrounds into the images or introducing color, texture, and adversarial changes in the background. We conduct extensive experiment to analyze the robustness of vision-based models against object-to-background context variations across diverse tasks. Code https://github.com/Muhammad-Huzaifaa/ObjectCompose.git

♻ ☆ Motion Generation from Fine-grained Textual Descriptions

Kunhang Li, Yansong Feng

The task of text2motion is to generate human motion sequences from given textual descriptions, where the model explores diverse mappings from natural language instructions to human body movements. While most existing works are confined to coarse-grained motion descriptions, e.g., "A man squats.", fine-grained descriptions specifying movements of relevant body parts are barely explored. Models trained with coarse-grained texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure to generate motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset specializing in fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with step-by-step instructions with pseudo-code compulsory checks. Accordingly, we design a new text2motion model, FineMotionDiffuse, making full use of fine-grained textual information. Our quantitative evaluation shows that FineMotionDiffuse trained on FineHumanML3D improves FID by a large margin of 0.38, compared with competitive baselines. According to the qualitative evaluation and case study, our model outperforms MotionDiffuse in generating spatially or chronologically composite motions, by learning the implicit mappings from fine-grained descriptions to the corresponding basic motions. We release our data at https://github.com/KunhangL/finemotiondiffuse.

♻ ☆ Towards Low-Energy Adaptive Personalization for Resource-Constrained Devices

Yushan Huang, Josh Millar, Yuxuan Long, Yuchen Zhao, Hamed Hadaddi

The personalization of machine learning (ML) models to address data drift is a significant challenge in the context of Internet of Things (IoT) applications. Presently, most approaches focus on fine-tuning either the full base model or its last few layers to adapt to new data, while often neglecting energy costs. However, various types of data drift exist, and fine-tuning the full base model or the last few layers may not result in optimal performance in certain scenarios. We propose Target Block Fine-Tuning (TBFT), a low-energy adaptive personalization framework designed for resource-constrained devices. We categorize data drift and personalization into three types: input-level, feature-level, and output-level. For each type, we fine-tune different blocks of the model to achieve optimal performance with reduced energy costs. Specifically, input-, feature-, and output-level correspond to fine-tuning the front, middle, and rear blocks of the model. We evaluate TBFT on a ResNet model, three datasets, three different training sizes, and a Raspberry Pi. Compared with the $Block Avg$, where each block is fine-tuned individually and their performance improvements are averaged, TBFT exhibits an improvement in model accuracy by an average of 15.30% whilst saving 41.57% energy consumption on average compared with full fine-tuning.

comment: Accepetd to The 4th Workshop on Machine Learning and Systems (EuroMLSys '24)

♻ ☆ FPT: Fine-grained Prompt Tuning for Parameter and Memory Efficient Fine Tuning in High-resolution Medical Image Classification

Yijin Huang, Pujin Cheng, Roger Tam, Xiaoying Tang

Parameter-efficient fine-tuning (PEFT) is proposed as a cost-effective way to transfer pre-trained models to downstream tasks, avoiding the high cost of updating entire large-scale pre-trained models (LPMs). In this work, we present Fine-grained Prompt Tuning (FPT), a novel PEFT method for medical image classification. FPT significantly reduces memory consumption compared to other PEFT methods, especially in high-resolution contexts. To achieve this, we first freeze the weights of the LPM and construct a learnable lightweight side network. The frozen LPM takes high-resolution images as input to extract fine-grained features, while the side network is fed low-resolution images to reduce memory usage. To allow the side network to access pre-trained knowledge, we introduce fine-grained prompts that summarize information from the LPM through a fusion module. Important tokens selection and preloading techniques are employed to further reduce training cost and memory requirements. We evaluate FPT on four medical datasets with varying sizes, modalities, and complexities. Experimental results demonstrate that FPT achieves comparable performance to fine-tuning the entire LPM while using only 1.8% of the learnable parameters and 13% of the memory costs of an encoder ViT-B model with a 512 x 512 input resolution.

♻ ☆ SegVol: Universal and Interactive Volumetric Medical Image Segmentation

Yuxin Du, Fan Bai, Tiejun Huang, Bo Zhao

Precise image segmentation provides clinical study with instructive information. Despite the remarkable progress achieved in medical image segmentation, there is still an absence of 3D foundation segmentation model that can segment a wide range of anatomical categories with easy user interaction. In this paper, we propose a 3D foundation segmentation model, named SegVol, supporting universal and interactive volumetric medical image segmentation. By scaling up training data to 90K unlabeled Computed Tomography (CT) volumes and 6K labeled CT volumes, this foundation model supports the segmentation of over 200 anatomical categories using semantic and spatial prompts. Extensive experiments on 10 internal validation tasks and 18 external validation tasks verify that SegVol outperforms the state of the art by a large margin. Through its capacity to provide precise volumetric segmentation across various anatomical categories, SegVol has the potential to accelerate advancements in medical imaging diagnosis and facilitate treatment optimization. The model and code are publicly available at: https://github.com/BAAI-DCAI/SegVol.

♻ ☆ DreamComposer: Controllable 3D Object Generation via Multi-View Conditions

Yunhan Yang, Yukun Huang, Xiaoyang Wu, Yuan-Chen Guo, Song-Hai Zhang, Hengshuang Zhao, Tong He, Xihui Liu

Utilizing pre-trained 2D large-scale generative models, recent works are capable of generating high-quality novel views from a single in-the-wild image. However, due to the lack of information from multiple views, these works encounter difficulties in generating controllable novel views. In this paper, we present DreamComposer, a flexible and scalable framework that can enhance existing view-aware diffusion models by injecting multi-view conditions. Specifically, DreamComposer first uses a view-aware 3D lifting module to obtain 3D representations of an object from multiple views. Then, it renders the latent features of the target view from 3D representations with the multi-view feature fusion module. Finally the target view features extracted from multi-view inputs are injected into a pre-trained diffusion model. Experiments show that DreamComposer is compatible with state-of-the-art diffusion models for zero-shot novel view synthesis, further enhancing them to generate high-fidelity novel view images with multi-view conditions, ready for controllable 3D object reconstruction and various other applications.

comment: Project Page: https://yhyang-myron.github.io/DreamComposer/

♻ ☆ Regularizing Self-supervised 3D Scene Flows with Surface Awareness and Cyclic Consistency

Patrik Vacek, David Hurych, Karel Zimmermann, Patrick Perez, Tomas Svoboda

Learning without supervision how to predict 3D scene flows from point clouds is essential to many perception systems. We propose a novel learning framework for this task which improves the necessary regularization. Relying on the assumption that scene elements are mostly rigid, current smoothness losses are built on the definition of ``rigid clusters" in the input point clouds. The definition of these clusters is challenging and has a significant impact on the quality of predicted flows. We introduce two new consistency losses that enlarge clusters while preventing them from spreading over distinct objects. In particular, we enforce \emph{temporal} consistency with a forward-backward cyclic loss and \emph{spatial} consistency by considering surface orientation similarity in addition to spatial proximity. The proposed losses are model-independent and can thus be used in a plug-and-play fashion to significantly improve the performance of existing models, as demonstrated on two most widely used architectures. We also showcase the effectiveness and generalization capability of our framework on four standard sensor-unique driving datasets, achieving state-of-the-art performance in 3D scene flow estimation. Our codes are available on https://github.com/ctu-vras/sac-flow.

♻ ☆ P2ANet: A Dataset and Benchmark for Dense Action Detection from Table Tennis Match Broadcasting Videos

Jiang Bian, Xuhong Li, Tao Wang, Qingzhong Wang, Jun Huang, Chen Liu, Jun Zhao, Feixiang Lu, Dejing Dou, Haoyi Xiong

While deep learning has been widely used for video analytics, such as video classification and action detection, dense action detection with fast-moving subjects from sports videos is still challenging. In this work, we release yet another sports video benchmark \TheName{} for \emph{\underline{P}}ing \emph{\underline{P}}ong-\emph{\underline{A}}ction detection, which consists of 2,721 video clips collected from the broadcasting videos of professional table tennis matches in World Table Tennis Championships and Olympiads. We work with a crew of table tennis professionals and referees on a specially designed annotation toolbox to obtain fine-grained action labels (in 14 classes) for every ping-pong action that appeared in the dataset, and formulate two sets of action detection problems -- \emph{action localization} and \emph{action recognition}. We evaluate a number of commonly-seen action recognition (e.g., TSM, TSN, Video SwinTransformer, and Slowfast) and action localization models (e.g., BSN, BSN++, BMN, TCANet), using \TheName{} for both problems, under various settings. These models can only achieve 48\% area under the AR-AN curve for localization and 82\% top-one accuracy for recognition since the ping-pong actions are dense with fast-moving subjects but broadcasting videos are with only 25 FPS. The results confirm that \TheName{} is still a challenging task and can be used as a special benchmark for dense action detection from videos.

♻ ☆ Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling CVPR 2024

Leon Sick, Dominik Engel, Pedro Hermosilla, Timo Ropinski

Traditionally, training neural networks to perform semantic segmentation required expensive human-made annotations. But more recently, advances in the field of unsupervised learning have made significant progress on this issue and towards closing the gap to supervised algorithms. To achieve this, semantic knowledge is distilled by learning to correlate randomly sampled features from images across an entire dataset. In this work, we build upon these advances by incorporating information about the structure of the scene into the training process through the use of depth information. We achieve this by (1) learning depth-feature correlation by spatially correlate the feature maps with the depth maps to induce knowledge about the structure of the scene and (2) implementing farthest-point sampling to more effectively select relevant features by utilizing 3D sampling techniques on depth information of the scene. Finally, we demonstrate the effectiveness of our technical contributions through extensive experimentation and present significant improvements in performance across multiple benchmark datasets.

comment: Accepted at CVPR 2024

♻ ☆ Diffusion Reflectance Map: Single-Image Stochastic Inverse Rendering of Illumination and Reflectance CVPR 2024

Yuto Enyo, Ko Nishino

Reflectance bounds the frequency spectrum of illumination in the object appearance. In this paper, we introduce the first stochastic inverse rendering method, which recovers the attenuated frequency spectrum of an illumination jointly with the reflectance of an object of known geometry from a single image. Our key idea is to solve this blind inverse problem in the reflectance map, an appearance representation invariant to the underlying geometry, by learning to reverse the image formation with a novel diffusion model which we refer to as the Diffusion Reflectance Map Network (DRMNet). Given an observed reflectance map converted and completed from the single input image, DRMNet generates a reflectance map corresponding to a perfect mirror sphere while jointly estimating the reflectance. The forward process can be understood as gradually filtering a natural illumination with lower and lower frequency reflectance and additive Gaussian noise. DRMNet learns to invert this process with two subnetworks, IllNet and RefNet, which work in concert towards this joint estimation. The network is trained on an extensive synthetic dataset and is demonstrated to generalize to real images, showing state-of-the-art accuracy on established datasets.

comment: to be published in CVPR 2024

♻ ☆ ProMamba: Prompt-Mamba for polyp segmentation

Jianhao Xie, Ruofan Liao, Ziang Zhang, Sida Yi, Yuesheng Zhu, Guibo Luo

Detecting polyps through colonoscopy is an important task in medical image segmentation, which provides significant assistance and reference value for clinical surgery. However, accurate segmentation of polyps is a challenging task due to two main reasons. Firstly, polyps exhibit various shapes and colors. Secondly, the boundaries between polyps and their normal surroundings are often unclear. Additionally, significant differences between different datasets lead to limited generalization capabilities of existing methods. To address these issues, we propose a segmentation model based on Prompt-Mamba, which incorporates the latest Vision-Mamba and prompt technologies. Compared to previous models trained on the same dataset, our model not only maintains high segmentation accuracy on the validation part of the same dataset but also demonstrates superior accuracy on unseen datasets, exhibiting excellent generalization capabilities. Notably, we are the first to apply the Vision-Mamba architecture to polyp segmentation and the first to utilize prompt technology in a polyp segmentation model. Our model efficiently accomplishes segmentation tasks, surpassing previous state-of-the-art methods by an average of 5% across six datasets. Furthermore, we have developed multiple versions of our model with scaled parameter counts, achieving better performance than previous models even with fewer parameters. Our code and trained weights will be released soon.

comment: 10 pages, 2 figures,3 tabels

♻ ☆ SocialCircle: Learning the Angle-based Social Interaction Representation for Pedestrian Trajectory Prediction CVPR 2024

Conghao Wong, Beihao Xia, Ziqian Zou, Yulong Wang, Xinge You

Analyzing and forecasting trajectories of agents like pedestrians and cars in complex scenes has become more and more significant in many intelligent systems and applications. The diversity and uncertainty in socially interactive behaviors among a rich variety of agents make this task more challenging than other deterministic computer vision tasks. Researchers have made a lot of efforts to quantify the effects of these interactions on future trajectories through different mathematical models and network structures, but this problem has not been well solved. Inspired by marine animals that localize the positions of their companions underwater through echoes, we build a new anglebased trainable social interaction representation, named SocialCircle, for continuously reflecting the context of social interactions at different angular orientations relative to the target agent. We validate the effect of the proposed SocialCircle by training it along with several newly released trajectory prediction models, and experiments show that the SocialCircle not only quantitatively improves the prediction performance, but also qualitatively helps better simulate social interactions when forecasting pedestrian trajectories in a way that is consistent with human intuitions.

comment: CVPR 2024 accepted

♻ ☆ Emotic Masked Autoencoder with Attention Fusion for Facial Expression Recognition

Bach Nguyen-Xuan, Thien Nguyen-Hoang, Nhu Tai-Do

Facial Expression Recognition (FER) is a critical task within computer vision with diverse applications across various domains. Addressing the challenge of limited FER datasets, which hampers the generalization capability of expression recognition models, is imperative for enhancing performance. Our paper presents an innovative approach integrating the MAE-Face self-supervised learning (SSL) method and Fusion Attention mechanism for expression classification, particularly showcased in the 6th Affective Behavior 32 pages harvmac; added references for section 5Analysis in-the-wild (ABAW) competition. Additionally, we propose preprocessing techniques to emphasize essential facial features, thereby enhancing model performance on both training and validation sets, notably demonstrated on the Aff-wild2 dataset.

comment: 6 pages; added references for section 1; corrected typo for email author

♻ ☆ Learning User Embeddings from Human Gaze for Personalised Saliency Prediction

Florian Strohm, Mihai Bâce, Andreas Bulling

Reusable embeddings of user behaviour have shown significant performance improvements for the personalised saliency prediction task. However, prior works require explicit user characteristics and preferences as input, which are often difficult to obtain. We present a novel method to extract user embeddings from pairs of natural images and corresponding saliency maps generated from a small amount of user-specific eye tracking data. At the core of our method is a Siamese convolutional neural encoder that learns the user embeddings by contrasting the image and personal saliency map pairs of different users. Evaluations on two public saliency datasets show that the generated embeddings have high discriminative power, are effective at refining universal saliency maps to the individual users, and generalise well across users and images. Finally, based on our model's ability to encode individual user characteristics, our work points towards other applications that can benefit from reusable embeddings of gaze behaviour.

♻ ☆ VRP-SAM: SAM with Visual Reference Prompt CVPR 2024

Yanpeng Sun, Jiahui Chen, Shan Zhang, Xinyu Zhang, Qiang Chen, Gang Zhang, Errui Ding, Jingdong Wang, Zechao Li

In this paper, we propose a novel Visual Reference Prompt (VRP) encoder that empowers the Segment Anything Model (SAM) to utilize annotated reference images as prompts for segmentation, creating the VRP-SAM model. In essence, VRP-SAM can utilize annotated reference images to comprehend specific objects and perform segmentation of specific objects in target image. It is note that the VRP encoder can support a variety of annotation formats for reference images, including \textbf{point}, \textbf{box}, \textbf{scribble}, and \textbf{mask}. VRP-SAM achieves a breakthrough within the SAM framework by extending its versatility and applicability while preserving SAM's inherent strengths, thus enhancing user-friendliness. To enhance the generalization ability of VRP-SAM, the VRP encoder adopts a meta-learning strategy. To validate the effectiveness of VRP-SAM, we conducted extensive empirical studies on the Pascal and COCO datasets. Remarkably, VRP-SAM achieved state-of-the-art performance in visual reference segmentation with minimal learnable parameters. Furthermore, VRP-SAM demonstrates strong generalization capabilities, allowing it to perform segmentation of unseen objects and enabling cross-domain segmentation. The source code and models will be available at \url{https://github.com/syp2ysy/VRP-SAM}

comment: Accepted by CVPR 2024; The camera-ready version

♻ ☆ SeFFeC: Semantic Facial Feature Control for Fine-grained Face Editing

Florian Strohm, Mihai Bâce, Markus Kaltenecker, Andreas Bulling

We propose Semantic Facial Feature Control (SeFFeC) - a novel method for fine-grained face shape editing. Our method enables the manipulation of human-understandable, semantic face features, such as nose length or mouth width, which are defined by different groups of facial landmarks. In contrast to existing methods, the use of facial landmarks enables precise measurement of the facial features, which then enables training SeFFeC without any manually annotated labels. SeFFeC consists of a transformer-based encoder network that takes a latent vector of a pre-trained generative model and a facial feature embedding as input, and learns to modify the latent vector to perform the desired face edit operation. To ensure that the desired feature measurement is changed towards the target value without altering uncorrelated features, we introduced a novel semantic face feature loss. Qualitative and quantitative results show that SeFFeC enables precise and fine-grained control of 23 facial features, some of which could not previously be controlled by other methods, without requiring manual annotations. Unlike existing methods, SeFFeC also provides deterministic control over the exact values of the facial features and more localised and disentangled face edits.

♻ ☆ Dual Prototype Attention for Unsupervised Video Object Segmentation CVPR 2024

Suhwan Cho, Minhyeok Lee, Seunghoon Lee, Dogyoon Lee, Heeseung Choi, Ig-Jae Kim, Sangyoun Lee

Unsupervised video object segmentation (VOS) aims to detect and segment the most salient object in videos. The primary techniques used in unsupervised VOS are 1) the collaboration of appearance and motion information; and 2) temporal fusion between different frames. This paper proposes two novel prototype-based attention mechanisms, inter-modality attention (IMA) and inter-frame attention (IFA), to incorporate these techniques via dense propagation across different modalities and frames. IMA densely integrates context information from different modalities based on a mutual refinement. IFA injects global context of a video to the query frame, enabling a full utilization of useful properties from multiple frames. Experimental results on public benchmark datasets demonstrate that our proposed approach outperforms all existing methods by a substantial margin. The proposed two components are also thoroughly validated via ablative study.

comment: CVPR 2024

♻ ☆ From Pretext to Purpose: Batch-Adaptive Self-Supervised Learning

Jiansong Zhang, Linlin Shen, Peizhong Liu

In recent years, self-supervised contrastive learning has emerged as a distinguished paradigm in the artificial intelligence landscape. It facilitates unsupervised feature learning through contrastive delineations at the instance level. However, crafting an effective self-supervised paradigm remains a pivotal challenge within this field. This paper delves into two crucial factors impacting self-supervised contrastive learning-bach size and pretext tasks, and from a data processing standpoint, proposes an adaptive technique of batch fusion. The proposed method, via dimensionality reduction and reconstruction of batch data, enables formerly isolated individual data to partake in intra-batch communication through the Embedding Layer. Moreover, it adaptively amplifies the self-supervised feature encoding capability as the training progresses. We conducted a linear classification test of this method based on the classic contrastive learning framework on ImageNet-1k. The empirical findings illustrate that our approach achieves state-of-the-art performance under equitable comparisons. Benefiting from its "plug-and-play" characteristics, we further explored other contrastive learning methods. On the ImageNet-100, compared to the original performance, the top1 has seen a maximum increase of 1.25%. We suggest that the proposed method may contribute to the advancement of data-driven self-supervised learning research, bringing a fresh perspective to this community.

comment: 14 pages, 2 figures, the code of this paper will be released soon

♻ ☆ LLaFS: When Large Language Models Meet Few-Shot Segmentation CVPR2024

Lanyun Zhu, Tianrun Chen, Deyi Ji, Jieping Ye, Jun Liu

This paper proposes LLaFS, the first attempt to leverage large language models (LLMs) in few-shot segmentation. In contrast to the conventional few-shot segmentation methods that only rely on the limited and biased information from the annotated support images, LLaFS leverages the vast prior knowledge gained by LLM as an effective supplement and directly uses the LLM to segment images in a few-shot manner. To enable the text-based LLM to handle image-related tasks, we carefully design an input instruction that allows the LLM to produce segmentation results represented as polygons, and propose a region-attribute table to simulate the human visual mechanism and provide multi-modal guidance. We also synthesize pseudo samples and use curriculum learning for pretraining to augment data and achieve better optimization. LLaFS achieves state-of-the-art results on multiple datasets, showing the potential of using LLMs for few-shot computer vision tasks.

comment: Accepted to CVPR2024

♻ ☆ EcoSense: Energy-Efficient Intelligent Sensing for In-Shore Ship Detection through Edge-Cloud Collaboration

Wenjun Huang, Hanning Chen, Yang Ni, Arghavan Rezvani, Sanggeon Yun, Sungheon Jeon, Eric Pedley, Mohsen Imani

Detecting marine objects inshore presents challenges owing to algorithmic intricacies and complexities in system deployment. We propose a difficulty-aware edge-cloud collaborative sensing system that splits the task into object localization and fine-grained classification. Objects are classified either at the edge or within the cloud, based on their estimated difficulty. The framework comprises a low-power device-tailored front-end model for object localization, classification, and difficulty estimation, along with a transformer-graph convolutional network-based back-end model for fine-grained classification. Our system demonstrates superior performance (mAP@0.5 +4.3%}) on widely used marine object detection datasets, significantly reducing both data transmission volume (by 95.43%) and energy consumption (by 72.7%}) at the system level. We validate the proposed system across various embedded system platforms and in real-world scenarios involving drone deployment.

♻ ☆ Vision Transformers with Hierarchical Attention

Yun Liu, Yu-Huan Wu, Guolei Sun, Le Zhang, Ajad Chhatkuli, Luc Van Gool

This paper tackles the high computational/space complexity associated with Multi-Head Self-Attention (MHSA) in vanilla vision transformers. To this end, we propose Hierarchical MHSA (H-MHSA), a novel approach that computes self-attention in a hierarchical fashion. Specifically, we first divide the input image into patches as commonly done, and each patch is viewed as a token. Then, the proposed H-MHSA learns token relationships within local patches, serving as local relationship modeling. Then, the small patches are merged into larger ones, and H-MHSA models the global dependencies for the small number of the merged tokens. At last, the local and global attentive features are aggregated to obtain features with powerful representation capacity. Since we only calculate attention for a limited number of tokens at each step, the computational load is reduced dramatically. Hence, H-MHSA can efficiently model global relationships among tokens without sacrificing fine-grained information. With the H-MHSA module incorporated, we build a family of Hierarchical-Attention-based Transformer Networks, namely HAT-Net. To demonstrate the superiority of HAT-Net in scene understanding, we conduct extensive experiments on fundamental vision tasks, including image classification, semantic segmentation, object detection, and instance segmentation. Therefore, HAT-Net provides a new perspective for vision transformers. Code and pretrained models are available at https://github.com/yun-liu/HAT-Net.

comment: Machine Intelligence Research (MIR), DOI: 10.1007/s11633-024-1393-8

♻ ☆ Domain-Aware Fine-Tuning: Enhancing Neural Network Adaptability

Seokhyeon Ha, Sunbeom Jung, Jungwoo Lee

Fine-tuning pre-trained neural network models has become a widely adopted approach across various domains. However, it can lead to the distortion of pre-trained feature extractors that already possess strong generalization capabilities. Mitigating feature distortion during adaptation to new target domains is crucial. Recent studies have shown promising results in handling feature distortion by aligning the head layer on in-distribution datasets before performing fine-tuning. Nonetheless, a significant limitation arises from the treatment of batch normalization layers during fine-tuning, leading to suboptimal performance. In this paper, we propose Domain-Aware Fine-Tuning (DAFT), a novel approach that incorporates batch normalization conversion and the integration of linear probing and fine-tuning. Our batch normalization conversion method effectively mitigates feature distortion by reducing modifications to the neural network during fine-tuning. Additionally, we introduce the integration of linear probing and fine-tuning to optimize the head layer with gradual adaptation of the feature extractor. By leveraging batch normalization layers and integrating linear probing and fine-tuning, our DAFT significantly mitigates feature distortion and achieves improved model performance on both in-distribution and out-of-distribution datasets. Extensive experiments demonstrate that our method outperforms other baseline methods, demonstrating its effectiveness in not only improving performance but also mitigating feature distortion.

♻ ☆ NeuS-PIR: Learning Relightable Neural Surface using Pre-Integrated Rendering

Shi Mao, Chenming Wu, Zhelun Shen, Yifan Wang, Dayan Wu, Liangjun Zhang

This paper presents a method, namely NeuS-PIR, for recovering relightable neural surfaces using pre-integrated rendering from multi-view images or video. Unlike methods based on NeRF and discrete meshes, our method utilizes implicit neural surface representation to reconstruct high-quality geometry, which facilitates the factorization of the radiance field into two components: a spatially varying material field and an all-frequency lighting representation. This factorization, jointly optimized using an adapted differentiable pre-integrated rendering framework with material encoding regularization, in turn addresses the ambiguity of geometry reconstruction and leads to better disentanglement and refinement of each scene property. Additionally, we introduced a method to distil indirect illumination fields from the learned representations, further recovering the complex illumination effect like inter-reflection. Consequently, our method enables advanced applications such as relighting, which can be seamlessly integrated with modern graphics engines. Qualitative and quantitative experiments have shown that NeuS-PIR outperforms existing methods across various tasks on both synthetic and real datasets. Source code is available at https://github.com/Sheldonmao/NeuSPIR

♻ ☆ Learning-based Axial Video Motion Magnification

Kwon Byung-Ki, Oh Hyun-Bin, Kim Jun-Seong, Hyunwoo Ha, Tae-Hyun Oh

Video motion magnification amplifies invisible small motions to be perceptible, which provides humans with a spatially dense and holistic understanding of small motions in the scene of interest. This is based on the premise that magnifying small motions enhances the legibility of motions. In the real world, however, vibrating objects often possess convoluted systems that have complex natural frequencies, modes, and directions. Existing motion magnification often fails to improve legibility since the intricate motions still retain complex characteristics even after being magnified, which may distract us from analyzing them. In this work, we focus on improving legibility by proposing a new concept, axial motion magnification, which magnifies decomposed motions along the user-specified direction. Axial motion magnification can be applied to various applications where motions of specific axes are critical, by providing simplified and easily readable motion information. To achieve this, we propose a novel Motion Separation Module that enables to disentangle and magnify the motion representation along axes of interest. Furthermore, we build a new synthetic training dataset for the axial motion magnification task. Our proposed method improves the legibility of resulting motions along certain axes by adding a new feature: user controllability. Axial motion magnification is a more generalized concept; thus, our method can be directly adapted to the generic motion magnification and achieves favorable performance against competing methods.

comment: main paper: 12 pages, supplementary: 10 pages, 20 figures, 1 table

♻ ☆ Decomposing Disease Descriptions for Enhanced Pathology Detection: A Multi-Aspect Vision-Language Pre-training Framework CVPR2024

Vu Minh Hieu Phan, Yutong Xie, Yuankai Qi, Lingqiao Liu, Liyang Liu, Bowen Zhang, Zhibin Liao, Qi Wu, Minh-Son To, Johan W. Verjans

Medical vision language pre-training (VLP) has emerged as a frontier of research, enabling zero-shot pathological recognition by comparing the query image with the textual descriptions for each disease. Due to the complex semantics of biomedical texts, current methods struggle to align medical images with key pathological findings in unstructured reports. This leads to the misalignment with the target disease's textual representation. In this paper, we introduce a novel VLP framework designed to dissect disease descriptions into their fundamental aspects, leveraging prior knowledge about the visual manifestations of pathologies. This is achieved by consulting a large language model and medical experts. Integrating a Transformer module, our approach aligns an input image with the diverse elements of a disease, generating aspect-centric image representations. By consolidating the matches from each aspect, we improve the compatibility between an image and its associated disease. Additionally, capitalizing on the aspect-oriented representations, we present a dual-head Transformer tailored to process known and unknown diseases, optimizing the comprehensive detection efficacy. Conducting experiments on seven downstream datasets, ours improves the accuracy of recent methods by up to 8.56% and 17.0% for seen and unseen categories, respectively. Our code is released at https://github.com/HieuPhan33/MAVL.

comment: Accepted at CVPR2024. Pre-print before final camera-ready version

♻ ☆ Lodge: A Coarse to Fine Diffusion Network for Long Dance Generation Guided by the Characteristic Dance Primitives CVPR2024

Ronghui Li, YuXiang Zhang, Yachao Zhang, Hongwen Zhang, Jie Guo, Yan Zhang, Yebin Liu, Xiu Li

We propose Lodge, a network capable of generating extremely long dance sequences conditioned on given music. We design Lodge as a two-stage coarse to fine diffusion architecture, and propose the characteristic dance primitives that possess significant expressiveness as intermediate representations between two diffusion models. The first stage is global diffusion, which focuses on comprehending the coarse-level music-dance correlation and production characteristic dance primitives. In contrast, the second-stage is the local diffusion, which parallelly generates detailed motion sequences under the guidance of the dance primitives and choreographic rules. In addition, we propose a Foot Refine Block to optimize the contact between the feet and the ground, enhancing the physical realism of the motion. Our approach can parallelly generate dance sequences of extremely long length, striking a balance between global choreographic patterns and local motion quality and expressiveness. Extensive experiments validate the efficacy of our method.

comment: Accepted by CVPR2024, Project page: https://li-ronghui.github.io/lodge

♻ ☆ LoCo: Locally Constrained Training-Free Layout-to-Image Synthesis

Peiang Zhao, Han Li, Ruiyang Jin, S. Kevin Zhou

Recent text-to-image diffusion models have reached an unprecedented level in generating high-quality images. However, their exclusive reliance on textual prompts often falls short in precise control of image compositions. In this paper, we propose LoCo, a training-free approach for layout-to-image Synthesis that excels in producing high-quality images aligned with both textual prompts and layout instructions. Specifically, we introduce a Localized Attention Constraint (LAC), leveraging semantic affinity between pixels in self-attention maps to create precise representations of desired objects and effectively ensure the accurate placement of objects in designated regions. We further propose a Padding Token Constraint (PTC) to leverage the semantic information embedded in previously neglected padding tokens, improving the consistency between object appearance and layout instructions. LoCo seamlessly integrates into existing text-to-image and layout-to-image models, enhancing their performance in spatial control and addressing semantic failures observed in prior methods. Extensive experiments showcase the superiority of our approach, surpassing existing state-of-the-art training-free layout-to-image methods both qualitatively and quantitatively across multiple benchmarks.

comment: Demo: https://huggingface.co/spaces/Pusheen/LoCo; Project page: https://momopusheen.github.io/LoCo/

♻ ☆ $\texttt{NePhi}$: Neural Deformation Fields for Approximately Diffeomorphic Medical Image Registration

Lin Tian, Hastings Greer, Raúl San José Estépar, Soumyadip Sengupta, Marc Niethammer

This work proposes NePhi, a generalizable neural deformation model which results in approximately diffeomorphic transformations. In contrast to the predominant voxel-based transformation fields used in learning-based registration approaches, NePhi represents deformations functionally, leading to great flexibility within the design space of memory consumption during training and inference, inference time, registration accuracy, as well as transformation regularity. Specifically, NePhi 1) requires less memory compared to voxel-based learning approaches, 2) improves inference speed by predicting latent codes, compared to current existing neural deformation based registration approaches that \emph{only} rely on optimization, 3) improves accuracy via instance optimization, and 4) shows excellent deformation regularity which is highly desirable for medical image registration. We demonstrate the performance of NePhi on a 2D synthetic dataset as well as for real 3D lung registration. Our results show that NePhi can match the accuracy of voxel-based representations in a single-resolution registration setting. For multi-resolution registration, our method matches the accuracy of current SOTA learning-based registration approaches with instance optimization while reducing memory requirements by a factor of five.

♻ ☆ A Novel Approach to Industrial Defect Generation through Blended Latent Diffusion Model with Online Adaptation

Hanxi Li, Zhengxun Zhang, Hao Chen, Lin Wu, Bo Li, Deyin Liu, Mingwen Wang

Effectively addressing the challenge of industrial Anomaly Detection (AD) necessitates an ample supply of defective samples, a constraint often hindered by their scarcity in industrial contexts. This paper introduces a novel algorithm designed to augment defective samples, thereby enhancing AD performance. The proposed method tailors the blended latent diffusion model for defect sample generation, employing a diffusion model to generate defective samples in the latent space. A feature editing process, controlled by a ``trimap" mask and text prompts, refines the generated samples. The image generation inference process is structured into three stages: a free diffusion stage, an editing diffusion stage, and an online decoder adaptation stage. This sophisticated inference strategy yields high-quality synthetic defective samples with diverse pattern variations, leading to significantly improved AD accuracies based on the augmented training set. Specifically, on the widely recognized MVTec AD dataset, the proposed method elevates the state-of-the-art (SOTA) performance of AD with augmented data by 1.5%, 1.9%, and 3.1% for AD metrics AP, IAP, and IAP90, respectively. The implementation code of this work can be found at the GitHub repository https://github.com/GrandpaXun242/AdaBLDM.git

comment: 13 pages,7 figures

♻ ☆ DiffCast: A Unified Framework via Residual Diffusion for Precipitation Nowcasting CVPR 2024

Demin Yu, Xutao Li, Yunming Ye, Baoquan Zhang, Chuyao Luo, Kuai Dai, Rui Wang, Xunlai Chen

Precipitation nowcasting is an important spatio-temporal prediction task to predict the radar echoes sequences based on current observations, which can serve both meteorological science and smart city applications. Due to the chaotic evolution nature of the precipitation systems, it is a very challenging problem. Previous studies address the problem either from the perspectives of deterministic modeling or probabilistic modeling. However, their predictions suffer from the blurry, high-value echoes fading away and position inaccurate issues. The root reason of these issues is that the chaotic evolutionary precipitation systems are not appropriately modeled. Inspired by the nature of the systems, we propose to decompose and model them from the perspective of global deterministic motion and local stochastic variations with residual mechanism. A unified and flexible framework that can equip any type of spatio-temporal models is proposed based on residual diffusion, which effectively tackles the shortcomings of previous methods. Extensive experimental results on four publicly available radar datasets demonstrate the effectiveness and superiority of the proposed framework, compared to state-of-the-art techniques. Our code is publicly available at https://github.com/DeminYu98/DiffCast.

comment: CVPR 2024; https://github.com/DeminYu98/DiffCast

♻ ☆ Diffusion Models Generate Images Like Painters: an Analytical Theory of Outline First, Details Later NeurIPS23

Binxu Wang, John J. Vastola

How do diffusion generative models convert pure noise into meaningful images? In a variety of pretrained diffusion models (including conditional latent space models like Stable Diffusion), we observe that the reverse diffusion process that underlies image generation has the following properties: (i) individual trajectories tend to be low-dimensional and resemble 2D `rotations'; (ii) high-variance scene features like layout tend to emerge earlier, while low-variance details tend to emerge later; and (iii) early perturbations tend to have a greater impact on image content than later perturbations. To understand these phenomena, we derive and study a closed-form solution to the probability flow ODE for a Gaussian distribution, which shows that the reverse diffusion state rotates towards a gradually-specified target on the image manifold. It also shows that generation involves first committing to an outline, and then to finer and finer details. We find that this solution accurately describes the initial phase of image generation for pretrained models, and can in principle be used to make image generation more efficient by skipping reverse diffusion steps. Finally, we use our solution to characterize the image manifold in Stable Diffusion. Our viewpoint reveals an unexpected similarity between generation by GANs and diffusion and provides a conceptual link between diffusion and image retrieval.

comment: 44 pages, 28 figures. A briefer version was presented at NeurIPS23 Workshop on Diffusion Models [arXiv:2311.10892]

♻ ☆ Confidence-Triggered Detection: Accelerating Real-time Tracking-by-detection Systems

Zhicheng Ding, Zhixin Lai, Siyang Li, Edward Wong

Real-time object tracking necessitates a delicate balance between speed and accuracy, a challenge exacerbated by the computational demands of deep learning methods. In this paper, we propose Confidence-Triggered Detection (CTD), an innovative approach that strategically bypasses object detection for frames closely resembling intermediate states, leveraging tracker confidence scores. CTD not only enhances tracking speed but also preserves accuracy, surpassing existing tracking algorithms. Through extensive evaluation across various tracker confidence thresholds, we identify an optimal trade-off between tracking speed and accuracy, providing crucial insights for parameter fine-tuning and enhancing CTD's practicality in real-world scenarios. Our experiments across diverse detection models underscore the robustness and versatility of the CTD framework, demonstrating its potential to enable real-time tracking in resource-constrained environments.

comment: 9 pages, 5 figures, 1 table

♻ ☆ Troika: Multi-Path Cross-Modal Traction for Compositional Zero-Shot Learning CVPR 2024

Siteng Huang, Biao Gong, Yutong Feng, Min Zhang, Yiliang Lv, Donglin Wang

Recent compositional zero-shot learning (CZSL) methods adapt pre-trained vision-language models (VLMs) by constructing trainable prompts only for composed state-object pairs. Relying on learning the joint representation of seen compositions, these methods ignore the explicit modeling of the state and object, thus limiting the exploitation of pre-trained knowledge and generalization to unseen compositions. With a particular focus on the universality of the solution, in this work, we propose a novel paradigm for CZSL models that establishes three identification branches (i.e., Multi-Path) to jointly model the state, object, and composition. The presented Troika is our implementation that aligns the branch-specific prompt representations with decomposed visual features. To calibrate the bias between semantically similar multi-modal representations, we further devise a Cross-Modal Traction module into Troika that shifts the prompt representation towards the current visual content. We conduct extensive experiments on three popular benchmarks, where our method significantly outperforms existing methods in both closed-world and open-world settings. The code will be available at https://github.com/bighuang624/Troika.

comment: CVPR 2024

♻ ☆ Joint Learning Neuronal Skeleton and Brain Circuit Topology with Permutation Invariant Encoders for Neuron Classification AAAI 2024

Minghui Liao, Guojia Wan, Bo Du

Determining the types of neurons within a nervous system plays a significant role in the analysis of brain connectomics and the investigation of neurological diseases. However, the efficiency of utilizing anatomical, physiological, or molecular characteristics of neurons is relatively low and costly. With the advancements in electron microscopy imaging and analysis techniques for brain tissue, we are able to obtain whole-brain connectome consisting neuronal high-resolution morphology and connectivity information. However, few models are built based on such data for automated neuron classification. In this paper, we propose NeuNet, a framework that combines morphological information of neurons obtained from skeleton and topological information between neurons obtained from neural circuit. Specifically, NeuNet consists of three components, namely Skeleton Encoder, Connectome Encoder, and Readout Layer. Skeleton Encoder integrates the local information of neurons in a bottom-up manner, with a one-dimensional convolution in neural skeleton's point data; Connectome Encoder uses a graph neural network to capture the topological information of neural circuit; finally, Readout Layer fuses the above two information and outputs classification results. We reprocess and release two new datasets for neuron classification task from volume electron microscopy(VEM) images of human brain cortex and Drosophila brain. Experiments on these two datasets demonstrated the effectiveness of our model with accuracy of 0.9169 and 0.9363, respectively. Code and data are available at: https://github.com/WHUminghui/NeuNet.

comment: Accepted by AAAI 2024

♻ ☆ Doubly Abductive Counterfactual Inference for Text-based Image Editing CVPR 2024

Xue Song, Jiequan Cui, Hanwang Zhang, Jingjing Chen, Richang Hong, Yu-Gang Jiang

We study text-based image editing (TBIE) of a single image by counterfactual inference because it is an elegant formulation to precisely address the requirement: the edited image should retain the fidelity of the original one. Through the lens of the formulation, we find that the crux of TBIE is that existing techniques hardly achieve a good trade-off between editability and fidelity, mainly due to the overfitting of the single-image fine-tuning. To this end, we propose a Doubly Abductive Counterfactual inference framework (DAC). We first parameterize an exogenous variable as a UNet LoRA, whose abduction can encode all the image details. Second, we abduct another exogenous variable parameterized by a text encoder LoRA, which recovers the lost editability caused by the overfitted first abduction. Thanks to the second abduction, which exclusively encodes the visual transition from post-edit to pre-edit, its inversion -- subtracting the LoRA -- effectively reverts pre-edit back to post-edit, thereby accomplishing the edit. Through extensive experiments, our DAC achieves a good trade-off between editability and fidelity. Thus, we can support a wide spectrum of user editing intents, including addition, removal, manipulation, replacement, style transfer, and facial change, which are extensively validated in both qualitative and quantitative evaluations. Codes are in https://github.com/xuesong39/DAC.

comment: Accepted by CVPR 2024

♻ ☆ Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models ICCV 2023

Songwei Ge, Seungjun Nah, Guilin Liu, Tyler Poon, Andrew Tao, Bryan Catanzaro, David Jacobs, Jia-Bin Huang, Ming-Yu Liu, Yogesh Balaji

Despite tremendous progress in generating high-quality images using diffusion models, synthesizing a sequence of animated frames that are both photorealistic and temporally coherent is still in its infancy. While off-the-shelf billion-scale datasets for image generation are available, collecting similar video data of the same scale is still challenging. Also, training a video diffusion model is computationally much more expensive than its image counterpart. In this work, we explore finetuning a pretrained image diffusion model with video data as a practical solution for the video synthesis task. We find that naively extending the image noise prior to video noise prior in video diffusion leads to sub-optimal performance. Our carefully designed video noise prior leads to substantially better performance. Extensive experimental validation shows that our model, Preserve Your Own Correlation (PYoCo), attains SOTA zero-shot text-to-video results on the UCF-101 and MSR-VTT benchmarks. It also achieves SOTA video generation quality on the small-scale UCF-101 benchmark with a $10\times$ smaller model using significantly less computation than the prior art.

comment: ICCV 2023. Project webpage: https://research.nvidia.com/labs/dir/pyoco

☆ SLEDGE: Synthesizing Simulation Environments for Driving Agents with Generative Models

Kashyap Chitta, Daniel Dauner, Andreas Geiger

SLEDGE is the first generative simulator for vehicle motion planning trained on real-world driving logs. Its core component is a learned model that is able to generate agent bounding boxes and lane graphs. The model's outputs serve as an initial state for traffic simulation. The unique properties of the entities to be generated for SLEDGE, such as their connectivity and variable count per scene, render the naive application of most modern generative models to this task non-trivial. Therefore, together with a systematic study of existing lane graph representations, we introduce a novel raster-to-vector autoencoder (RVAE). It encodes agents and the lane graph into distinct channels in a rasterized latent map. This facilitates both lane-conditioned agent generation and combined generation of lanes and agents with a Diffusion Transformer. Using generated entities in SLEDGE enables greater control over the simulation, e.g. upsampling turns or increasing traffic density. Further, SLEDGE can support 500m long routes, a capability not found in existing data-driven simulators like nuPlan. It presents new challenges for planning algorithms, evidenced by failure rates of over 40% for PDM, the winner of the 2023 nuPlan challenge, when tested on hard routes and dense traffic generated by our model. Compared to nuPlan, SLEDGE requires 500$\times$ less storage to set up (<4GB), making it a more accessible option and helping with democratizing future research in this field.

☆ MAGIS: LLM-Based Multi-Agent Framework for GitHub Issue Resolution

Wei Tao, Yucheng Zhou, Wenqiang Zhang, Yu Cheng

In software evolution, resolving the emergent issues within GitHub repositories is a complex challenge that involves not only the incorporation of new code but also the maintenance of existing functionalities. Large Language Models (LLMs) have shown promise in code generation and understanding but face difficulties in code change, particularly at the repository level. To overcome these challenges, we empirically study the reason why LLMs mostly fail to resolve GitHub issues and analyze some impact factors. Motivated by the empirical findings, we propose a novel LLM-based Multi-Agent framework for GitHub Issue reSolution, MAGIS, consisting of four kinds of agents customized for the software evolution: Manager, Repository Custodian, Developer, and Quality Assurance Engineer agents. This framework leverages the collaboration of various agents in the planning and coding process to unlock the potential of LLMs to resolve GitHub issues. In experiments, we employ the SWE-bench benchmark to compare MAGIS with popular LLMs, including GPT-3.5, GPT-4, and Claude-2. MAGIS can resolve 13.94% GitHub issues, which significantly outperforms the baselines. Specifically, MAGIS achieves an eight-fold increase in resolved ratio over the direct application of GPT-4, the based LLM of our method. We also analyze the factors for improving GitHub issue resolution rates, such as line location, task allocation, etc.

comment: work in progress

☆ AID: Attention Interpolation of Text-to-Image Diffusion

Qiyuan He, Jinghao Wang, Ziwei Liu, Angela Yao

Conditional diffusion models can create unseen images in various settings, aiding image interpolation. Interpolation in latent spaces is well-studied, but interpolation with specific conditions like text or poses is less understood. Simple approaches, such as linear interpolation in the space of conditions, often result in images that lack consistency, smoothness, and fidelity. To that end, we introduce a novel training-free technique named Attention Interpolation via Diffusion (AID). Our key contributions include 1) proposing an inner/outer interpolated attention layer; 2) fusing the interpolated attention with self-attention to boost fidelity; and 3) applying beta distribution to selection to increase smoothness. We also present a variant, Prompt-guided Attention Interpolation via Diffusion (PAID), that considers interpolation as a condition-dependent generative process. This method enables the creation of new images with greater consistency, smoothness, and efficiency, and offers control over the exact path of interpolation. Our approach demonstrates effectiveness for conceptual and spatial interpolation. Code and demo are available at https://github.com/QY-H00/attention-interpolation-diffusion.

☆ LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning

Rui Pan, Xiang Liu, Shizhe Diao, Renjie Pi, Jipeng Zhang, Chi Han, Tong Zhang

The machine learning community has witnessed impressive advancements since the first appearance of large language models (LLMs), yet their huge memory consumption has become a major roadblock to large-scale training. Parameter Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA) have been proposed to alleviate this problem, but their performance still fails to match full parameter training in most large-scale fine-tuning settings. Attempting to complement this deficiency, we investigate layerwise properties of LoRA on fine-tuning tasks and observe an uncommon skewness of weight norms across different layers. Utilizing this key observation, a surprisingly simple training strategy is discovered, which outperforms both LoRA and full parameter training in a wide range of settings with memory costs as low as LoRA. We name it Layerwise Importance Sampled AdamW (LISA), a promising alternative for LoRA, which applies the idea of importance sampling to different layers in LLMs and randomly freeze most middle layers during optimization. Experimental results show that with similar or less GPU memory consumption, LISA surpasses LoRA or even full parameter tuning in downstream fine-tuning tasks, where LISA consistently outperforms LoRA by over $11\%$-$37\%$ in terms of MT-Bench scores. On large models, specifically LLaMA-2-70B, LISA achieves on-par or better performance than LoRA on MT-Bench, GSM8K, and PubMedQA, demonstrating its effectiveness across different domains.

☆ AgentStudio: A Toolkit for Building General Virtual Agents

Longtao Zheng, Zhiyuan Huang, Zhenghai Xue, Xinrun Wang, Bo An, Shuicheng Yan

Creating autonomous virtual agents capable of using arbitrary software on any digital device remains a major challenge for artificial intelligence. Two key obstacles hinder progress: insufficient infrastructure for building virtual agents in real-world environments, and the need for in-the-wild evaluation of fundamental agent abilities. To address this, we introduce AgentStudio, an online, realistic, and multimodal toolkit that covers the entire lifecycle of agent development. This includes environment setups, data collection, agent evaluation, and visualization. The observation and action spaces are highly generic, supporting both function calling and human-computer interfaces. This versatility is further enhanced by AgentStudio's graphical user interfaces, which allow efficient development of datasets and benchmarks in real-world settings. To illustrate, we introduce a visual grounding dataset and a real-world benchmark suite, both created with our graphical interfaces. Furthermore, we present several actionable insights derived from AgentStudio, e.g., general visual grounding, open-ended tool creation, learning from videos, etc. We have open-sourced the environments, datasets, benchmarks, and interfaces to promote research towards developing general virtual agents for the future.

☆ CMP: Cooperative Motion Prediction with Multi-Agent Communication

Zhuoyuan Wu, Yuping Wang, Hengbo Ma, Zhaowei Li, Hang Qiu, Jiachen Li

The confluence of the advancement of Autonomous Vehicles (AVs) and the maturity of Vehicle-to-Everything (V2X) communication has enabled the capability of cooperative connected and automated vehicles (CAVs). Building on top of cooperative perception, this paper explores the feasibility and effectiveness of cooperative motion prediction. Our method, CMP, takes LiDAR signals as input to enhance tracking and prediction capabilities. Unlike previous work that focuses separately on either cooperative perception or motion prediction, our framework, to the best of our knowledge, is the first to address the unified problem where CAVs share information in both perception and prediction modules. Incorporated into our design is the unique capability to tolerate realistic V2X bandwidth limitations and transmission delays, while dealing with bulky perception representations. We also propose a prediction aggregation module, which unifies the predictions obtained by different CAVs and generates the final prediction. Through extensive experiments and ablation studies, we demonstrate the effectiveness of our method in cooperative perception, tracking, and motion prediction tasks. In particular, CMP reduces the average prediction error by 17.2\% with fewer missing detections compared with the no cooperation setting. Our work marks a significant step forward in the cooperative capabilities of CAVs, showcasing enhanced performance in complex scenarios.

★ Hierarchical Multi-label Classification for Fine-level Event Extraction from Aviation Accident Reports

Xinyu Zhao, Hao Yan, Yongming Liu

A large volume of accident reports is recorded in the aviation domain, which greatly values improving aviation safety. To better use those reports, we need to understand the most important events or impact factors according to the accident reports. However, the increasing number of accident reports requires large efforts from domain experts to label those reports. In order to make the labeling process more efficient, many researchers have started developing algorithms to identify the underlying events from accident reports automatically. This article argues that we can identify the events more accurately by leveraging the event taxonomy. More specifically, we consider the problem a hierarchical classification task where we first identify the coarse-level information and then predict the fine-level information. We achieve this hierarchical classification process by incorporating a novel hierarchical attention module into BERT. To further utilize the information from event taxonomy, we regularize the proposed model according to the relationship and distribution among labels. The effectiveness of our framework is evaluated with the data collected by National Transportation Safety Board (NTSB). It has been shown that fine-level prediction accuracy is highly improved, and the regularization term can be beneficial to the rare event identification problem.

comment: Accepted in INFORMS Journal of Data Science

☆ Image-based Novel Fault Detection with Deep Learning Classifiers using Hierarchical Labels

Nurettin Sergin, Jiayu Huang, Tzyy-Shuh Chang, Hao Yan

One important characteristic of modern fault classification systems is the ability to flag the system when faced with previously unseen fault types. This work considers the unknown fault detection capabilities of deep neural network-based fault classifiers. Specifically, we propose a methodology on how, when available, labels regarding the fault taxonomy can be used to increase unknown fault detection performance without sacrificing model performance. To achieve this, we propose to utilize soft label techniques to improve the state-of-the-art deep novel fault detection techniques during the training process and novel hierarchically consistent detection statistics for online novel fault detection. Finally, we demonstrated increased detection performance on novel fault detection in inspection images from the hot steel rolling process, with results well replicated across multiple scenarios and baseline detection methods.

comment: Accepted in IISE Transaction

☆ Addressing Social Misattributions of Large Language Models: An HCXAI-based Approach

Andrea Ferrario, Alberto Termine, Alessandro Facchini

Human-centered explainable AI (HCXAI) advocates for the integration of social aspects into AI explanations. Central to the HCXAI discourse is the Social Transparency (ST) framework, which aims to make the socio-organizational context of AI systems accessible to their users. In this work, we suggest extending the ST framework to address the risks of social misattributions in Large Language Models (LLMs), particularly in sensitive areas like mental health. In fact LLMs, which are remarkably capable of simulating roles and personas, may lead to mismatches between designers' intentions and users' perceptions of social attributes, risking to promote emotional manipulation and dangerous behaviors, cases of epistemic injustice, and unwarranted trust. To address these issues, we propose enhancing the ST framework with a fifth 'W-question' to clarify the specific social attributions assigned to LLMs by its designers and users. This addition aims to bridge the gap between LLM capabilities and user perceptions, promoting the ethically responsible development and use of LLM-based technology.

comment: Extended version of the manuscript accepted for the ACM CHI Workshop on Human-Centered Explainable AI 2024 (HCXAI24)

☆ Climate Downscaling: A Deep-Learning Based Super-resolution Model of Precipitation Data with Attention Block and Skip Connections

Chia-Hao Chiang, Zheng-Han Huang, Liwen Liu, Hsin-Chien Liang, Yi-Chi Wang, Wan-Ling Tseng, Chao Wang, Che-Ta Chen, Ko-Chih Wang

Human activities accelerate consumption of fossil fuels and produce greenhouse gases, resulting in urgent issues today: global warming and the climate change. These indirectly cause severe natural disasters, plenty of lives suffering and huge losses of agricultural properties. To mitigate impacts on our lands, scientists are developing renewable, reusable, and clean energies and climatologists are trying to predict the extremes. Meanwhile, governments are publicizing resource-saving policies for a more eco-friendly society and arousing environment awareness. One of the most influencing factors is the precipitation, bringing condensed water vapor onto lands. Water resources are the most significant but basic needs in society, not only supporting our livings, but also economics. In Taiwan, although the average annual precipitation is up to 2,500 millimeter (mm), the water allocation for each person is lower than the global average due to drastically geographical elevation changes and uneven distribution through the year. Thus, it is crucial to track and predict the rainfall to make the most use of it and to prevent the floods. However, climate models have limited resolution and require intensive computational power for local-scale use. Therefore, we proposed a deep convolutional neural network with skip connections, attention blocks, and auxiliary data concatenation, in order to downscale the low-resolution precipitation data into high-resolution one. Eventually, we compare with other climate downscaling methods and show better performance in metrics of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Pearson Correlation, structural similarity index (SSIM), and forecast indicators.

★ Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation

Abdelrhman Werby, Chenguang Huang, Martin Büchner, Abhinav Valada, Wolfram Burgard

Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features. While these maps allow for the prediction of point-wise saliency maps when queried for a certain language concept, large-scale environments and abstract queries beyond the object level still pose a considerable hurdle, ultimately limiting language-grounded robotic navigation. In this work, we present HOV-SG, a hierarchical open-vocabulary 3D scene graph mapping approach for language-grounded robot navigation. Leveraging open-vocabulary vision foundation models, we first obtain state-of-the-art open-vocabulary segment-level maps in 3D and subsequently construct a 3D scene graph hierarchy consisting of floor, room, and object concepts, each enriched with open-vocabulary features. Our approach is able to represent multi-story buildings and allows robotic traversal of those using a cross-floor Voronoi graph. HOV-SG is evaluated on three distinct datasets and surpasses previous baselines in open-vocabulary semantic accuracy on the object, room, and floor level while producing a 75% reduction in representation size compared to dense open-vocabulary maps. In order to prove the efficacy and generalization capabilities of HOV-SG, we showcase successful long-horizon language-conditioned robot navigation within real-world multi-storage environments. We provide code and trial video data at http://hovsg.github.io/.

comment: Code and video are available at http://hovsg.github.io/

☆ ReMamber: Referring Image Segmentation with Mamba Twister

Yuhuan Yang, Chaofan Ma, Jiangchao Yao, Zhun Zhong, Ya Zhang, Yanfeng Wang

Referring Image Segmentation (RIS) leveraging transformers has achieved great success on the interpretation of complex visual-language tasks. However, the quadratic computation cost makes it resource-consuming in capturing long-range visual-language dependencies. Fortunately, Mamba addresses this with efficient linear complexity in processing. However, directly applying Mamba to multi-modal interactions presents challenges, primarily due to inadequate channel interactions for the effective fusion of multi-modal data. In this paper, we propose ReMamber, a novel RIS architecture that integrates the power of Mamba with a multi-modal Mamba Twister block. The Mamba Twister explicitly models image-text interaction, and fuses textual and visual features through its unique channel and spatial twisting mechanism. We achieve the state-of-the-art on three challenging benchmarks. Moreover, we conduct thorough analyses of ReMamber and discuss other fusion designs using Mamba. These provide valuable perspectives for future research.

☆ DiffH2O: Diffusion-Based Synthesis of Hand-Object Interactions from Textual Descriptions

Sammy Christen, Shreyas Hampali, Fadime Sener, Edoardo Remelli, Tomas Hodan, Eric Sauser, Shugao Ma, Bugra Tekin

Generating natural hand-object interactions in 3D is challenging as the resulting hand and object motions are expected to be physically plausible and semantically meaningful. Furthermore, generalization to unseen objects is hindered by the limited scale of available hand-object interaction datasets. We propose DiffH2O, a novel method to synthesize realistic, one or two-handed object interactions from provided text prompts and geometry of the object. The method introduces three techniques that enable effective learning from limited data. First, we decompose the task into a grasping stage and a text-based interaction stage and use separate diffusion models for each. In the grasping stage, the model only generates hand motions, whereas in the interaction phase both hand and object poses are synthesized. Second, we propose a compact representation that tightly couples hand and object poses. Third, we propose two different guidance schemes to allow more control of the generated motions: grasp guidance and detailed textual guidance. Grasp guidance takes a single target grasping pose and guides the diffusion model to reach this grasp at the end of the grasping stage, which provides control over the grasping pose. Given a grasping motion from this stage, multiple different actions can be prompted in the interaction phase. For textual guidance, we contribute comprehensive text descriptions to the GRAB dataset and show that they enable our method to have more fine-grained control over hand-object interactions. Our quantitative and qualitative evaluation demonstrates that the proposed method outperforms baseline methods and leads to natural hand-object motions. Moreover, we demonstrate the practicality of our framework by utilizing a hand pose estimate from an off-the-shelf pose estimator for guidance, and then sampling multiple different actions in the interaction stage.

comment: Project Page: https://diffh2o.github.io/

☆ On the Computational Complexity of Stackelberg Planning and Meta-Operator Verification: Technical Report

Gregor Behnke, Marcel Steinmetz

Stackelberg planning is a recently introduced single-turn two-player adversarial planning model, where two players are acting in a joint classical planning task, the objective of the first player being hampering the second player from achieving its goal. This places the Stackelberg planning problem somewhere between classical planning and general combinatorial two-player games. But, where exactly? All investigations of Stackelberg planning so far focused on practical aspects. We close this gap by conducting the first theoretical complexity analysis of Stackelberg planning. We show that in general Stackelberg planning is actually no harder than classical planning. Under a polynomial plan-length restriction, however, Stackelberg planning is a level higher up in the polynomial complexity hierarchy, suggesting that compilations into classical planning come with a worst-case exponential plan-length increase. In attempts to identify tractable fragments, we further study its complexity under various planning task restrictions, showing that Stackelberg planning remains intractable where classical planning is not. We finally inspect the complexity of meta-operator verification, a problem that has been recently connected to Stackelberg planning.

comment: Presented at ICAPS24

☆ Accelerating Radio Spectrum Regulation Workflows with Large Language Models (LLMs)

Amir Ghasemi, Paul Guinand

Wireless spectrum regulation is a complex and demanding process due to the rapid pace of technological progress, increasing demand for spectrum, and a multitude of stakeholders with potentially conflicting interests, alongside significant economic implications. To navigate this, regulators must engage effectively with all parties, keep pace with global technology trends, conduct technical evaluations, issue licenses in a timely manner, and comply with various legal and policy frameworks. In light of these challenges, this paper demonstrates example applications of Large Language Models (LLMs) to expedite spectrum regulatory processes. We explore various roles that LLMs can play in this context while identifying some of the challenges to address. The paper also offers practical case studies and insights, with appropriate experiments, highlighting the transformative potential of LLMs in spectrum management.

☆ D-PAD: Deep-Shallow Multi-Frequency Patterns Disentangling for Time Series Forecasting

Xiaobing Yuan, Ling Chen

In time series forecasting, effectively disentangling intricate temporal patterns is crucial. While recent works endeavor to combine decomposition techniques with deep learning, multiple frequencies may still be mixed in the decomposed components, e.g., trend and seasonal. Furthermore, frequency domain analysis methods, e.g., Fourier and wavelet transforms, have limitations in resolution in the time domain and adaptability. In this paper, we propose D-PAD, a deep-shallow multi-frequency patterns disentangling neural network for time series forecasting. Specifically, a multi-component decomposing (MCD) block is introduced to decompose the series into components with different frequency ranges, corresponding to the "shallow" aspect. A decomposition-reconstruction-decomposition (D-R-D) module is proposed to progressively extract the information of frequencies mixed in the components, corresponding to the "deep" aspect. After that, an interaction and fusion (IF) module is used to further analyze the components. Extensive experiments on seven real-world datasets demonstrate that D-PAD achieves the state-of-the-art performance, outperforming the best baseline by an average of 9.48% and 7.15% in MSE and MAE, respectively.

☆ Evaluating the Efficacy of Prompt-Engineered Large Multimodal Models Versus Fine-Tuned Vision Transformers in Image-Based Security Applications

Fouad Trad, Ali Chehab

The success of Large Language Models (LLMs) has led to a parallel rise in the development of Large Multimodal Models (LMMs), such as Gemini-pro, which have begun to transform a variety of applications. These sophisticated multimodal models are designed to interpret and analyze complex data, integrating both textual and visual information on a scale previously unattainable, opening new avenues for a range of applications. This paper investigates the applicability and effectiveness of prompt-engineered Gemini-pro LMMs versus fine-tuned Vision Transformer (ViT) models in addressing critical security challenges. We focus on two distinct tasks: a visually evident task of detecting simple triggers, such as small squares in images, indicative of potential backdoors, and a non-visually evident task of malware classification through visual representations. Our results highlight a significant divergence in performance, with Gemini-pro falling short in accuracy and reliability when compared to fine-tuned ViT models. The ViT models, on the other hand, demonstrate exceptional accuracy, achieving near-perfect performance on both tasks. This study not only showcases the strengths and limitations of prompt-engineered LMMs in cybersecurity applications but also emphasizes the unmatched efficacy of fine-tuned ViT models for precise and dependable tasks.

☆ SciCapenter: Supporting Caption Composition for Scientific Figures with Machine-Generated Captions and Ratings

Ting-Yao Hsu, Chieh-Yang Huang, Shih-Hong Huang, Ryan Rossi, Sungchul Kim, Tong Yu, C. Lee Giles, Ting-Hao K. Huang

Crafting effective captions for figures is important. Readers heavily depend on these captions to grasp the figure's message. However, despite a well-developed set of AI technologies for figures and captions, these have rarely been tested for usefulness in aiding caption writing. This paper introduces SciCapenter, an interactive system that puts together cutting-edge AI technologies for scientific figure captions to aid caption composition. SciCapenter generates a variety of captions for each figure in a scholarly article, providing scores and a comprehensive checklist to assess caption quality across multiple critical aspects, such as helpfulness, OCR mention, key takeaways, and visual properties reference. Users can directly edit captions in SciCapenter, resubmit for revised evaluations, and iteratively refine them. A user study with Ph.D. students indicates that SciCapenter significantly lowers the cognitive load of caption writing. Participants' feedback further offers valuable design insights for future systems aiming to enhance caption writing.

comment: CHI EA '24: Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems

☆ Towards a FAIR Documentation of Workflows and Models in Applied Mathematics

Marco Reidelbach, Björn Schembera, Marcus Weber

Modeling-Simulation-Optimization workflows play a fundamental role in applied mathematics. The Mathematical Research Data Initiative, MaRDI, responded to this by developing a FAIR and machine-interpretable template for a comprehensive documentation of such workflows. MaRDMO, a Plugin for the Research Data Management Organiser, enables scientists from diverse fields to document and publish their workflows on the MaRDI Portal seamlessly using the MaRDI template. Central to these workflows are mathematical models. MaRDI addresses them with the MathModDB ontology, offering a structured formal model description. Here, we showcase the interaction between MaRDMO and the MathModDB Knowledge Graph through an algebraic modeling workflow from the Digital Humanities. This demonstration underscores the versatility of both services beyond their original numerical domain.

☆ SciNews: From Scholarly Complexities to Public Narratives -- A Dataset for Scientific News Report Generation LREC

Dongqi Pu, Yifan Wang, Jia Loy, Vera Demberg

Scientific news reports serve as a bridge, adeptly translating complex research articles into reports that resonate with the broader public. The automated generation of such narratives enhances the accessibility of scholarly insights. In this paper, we present a new corpus to facilitate this paradigm development. Our corpus comprises a parallel compilation of academic publications and their corresponding scientific news reports across nine disciplines. To demonstrate the utility and reliability of our dataset, we conduct an extensive analysis, highlighting the divergences in readability and brevity between scientific news narratives and academic manuscripts. We benchmark our dataset employing state-of-the-art text generation models. The evaluation process involves both automatic and human evaluation, which lays the groundwork for future explorations into the automated generation of scientific news reports. The dataset and code related to this work are available at https://dongqi.me/projects/SciNews.

comment: LREC-COLING 2024 Main Conference Paper

★ DataCook: Crafting Anti-Adversarial Examples for Healthcare Data Copyright Protection

Sihan Shang, Jiancheng Yang, Zhenglong Sun, Pascal Fua

In the realm of healthcare, the challenges of copyright protection and unauthorized third-party misuse are increasingly significant. Traditional methods for data copyright protection are applied prior to data distribution, implying that models trained on these data become uncontrollable. This paper introduces a novel approach, named DataCook, designed to safeguard the copyright of healthcare data during the deployment phase. DataCook operates by "cooking" the raw data before distribution, enabling the development of models that perform normally on this processed data. However, during the deployment phase, the original test data must be also "cooked" through DataCook to ensure normal model performance. This process grants copyright holders control over authorization during the deployment phase. The mechanism behind DataCook is by crafting anti-adversarial examples (AntiAdv), which are designed to enhance model confidence, as opposed to standard adversarial examples (Adv) that aim to confuse models. Similar to Adv, AntiAdv introduces imperceptible perturbations, ensuring that the data processed by DataCook remains easily understandable. We conducted extensive experiments on MedMNIST datasets, encompassing both 2D/3D data and the high-resolution variants. The outcomes indicate that DataCook effectively meets its objectives, preventing models trained on AntiAdv from analyzing unauthorized data effectively, without compromising the validity and accuracy of the data in legitimate scenarios. Code and data are available at https://github.com/MedMNIST/DataCook.

☆ Using Stratified Sampling to Improve LIME Image Explanations

Muhammad Rashid, Elvio G. Amparore, Enrico Ferrari, Damiano Verda

We investigate the use of a stratified sampling approach for LIME Image, a popular model-agnostic explainable AI method for computer vision tasks, in order to reduce the artifacts generated by typical Monte Carlo sampling. Such artifacts are due to the undersampling of the dependent variable in the synthetic neighborhood around the image being explained, which may result in inadequate explanations due to the impossibility of fitting a linear regressor on the sampled data. We then highlight a connection with the Shapley theory, where similar arguments about undersampling and sample relevance were suggested in the past. We derive all the formulas and adjustment factors required for an unbiased stratified sampling estimator. Experiments show the efficacy of the proposed approach.

☆ All-in-One: Heterogeneous Interaction Modeling for Cold-Start Rating Prediction

Shuheng Fang, Kangfei Zhao, Yu Rong, Zhixun Li, Jeffrey Xu Yu

Cold-start rating prediction is a fundamental problem in recommender systems that has been extensively studied. Many methods have been proposed that exploit explicit relations among existing data, such as collaborative filtering, social recommendations and heterogeneous information network, to alleviate the data insufficiency issue for cold-start users and items. However, the explicit relations constructed based on data between different roles may be unreliable and irrelevant, which limits the performance ceiling of the specific recommendation task. Motivated by this, in this paper, we propose a flexible framework dubbed heterogeneous interaction rating network (HIRE). HIRE dose not solely rely on the pre-defined interaction pattern or the manually constructed heterogeneous information network. Instead, we devise a Heterogeneous Interaction Module (HIM) to jointly model the heterogeneous interactions and directly infer the important interactions via the observed data. In the experiments, we evaluate our model under three cold-start settings on three real-world datasets. The experimental results show that HIRE outperforms other baselines by a large margin. Furthermore, we visualize the inferred interactions of HIRE to confirm the contribution of our model.

comment: 14 pages, 9 figures

☆ Out-of-distribution Rumor Detection via Test-Time Adaptation

Xiang Tao, Mingqing Zhang, Qiang Liu, Shu Wu, Liang Wang

Due to the rapid spread of rumors on social media, rumor detection has become an extremely important challenge. Existing methods for rumor detection have achieved good performance, as they have collected enough corpus from the same data distribution for model training. However, significant distribution shifts between the training data and real-world test data occur due to differences in news topics, social media platforms, languages and the variance in propagation scale caused by news popularity. This leads to a substantial decline in the performance of these existing methods in Out-Of-Distribution (OOD) situations. To address this problem, we propose a simple and efficient method named Test-time Adaptation for Rumor Detection under distribution shifts (TARD). This method models the propagation of news in the form of a propagation graph, and builds propagation graph test-time adaptation framework, enhancing the model's adaptability and robustness when facing OOD problems. Extensive experiments conducted on two group datasets collected from real-world social platforms demonstrate that our framework outperforms the state-of-the-art methods in performance.

☆ Tiny Models are the Computational Saver for Large Models

Qingyuan Wang, Barry Cardiff, Antoine Frappé, Benoit Larras, Deepu John

This paper introduces TinySaver, an early-exit-like dynamic model compression approach which employs tiny models to substitute large models adaptively. Distinct from traditional compression techniques, dynamic methods like TinySaver can leverage the difficulty differences to allow certain inputs to complete their inference processes early, thereby conserving computational resources. Most existing early exit designs are implemented by attaching additional network branches to the model's backbone. Our study, however, reveals that completely independent tiny models can replace a substantial portion of the larger models' job with minimal impact on performance. Employing them as the first exit can remarkably enhance computational efficiency. By searching and employing the most appropriate tiny model as the computational saver for a given large model, the proposed approaches work as a novel and generic method to model compression. This finding will help the research community in exploring new compression methods to address the escalating computational demands posed by rapidly evolving AI models. Our evaluation of this approach in ImageNet-1k classification demonstrates its potential to reduce the number of compute operations by up to 90%, with only negligible losses in performance, across various modern vision models. The code of this work will be available.

☆ Optimization-based Prompt Injection Attack to LLM-as-a-Judge

Jiawen Shi, Zenghui Yuan, Yinuo Liu, Yue Huang, Pan Zhou, Lichao Sun, Neil Zhenqiang Gong

LLM-as-a-Judge is a novel solution that can assess textual information with large language models (LLMs). Based on existing research studies, LLMs demonstrate remarkable performance in providing a compelling alternative to traditional human assessment. However, the robustness of these systems against prompt injection attacks remains an open question. In this work, we introduce JudgeDeceiver, a novel optimization-based prompt injection attack tailored to LLM-as-a-Judge. Our method formulates a precise optimization objective for attacking the decision-making process of LLM-as-a-Judge and utilizes an optimization algorithm to efficiently automate the generation of adversarial sequences, achieving targeted and effective manipulation of model evaluations. Compared to handcraft prompt injection attacks, our method demonstrates superior efficacy, posing a significant challenge to the current security paradigms of LLM-based judgment systems. Through extensive experiments, we showcase the capability of JudgeDeceiver in altering decision outcomes across various cases, highlighting the vulnerability of LLM-as-a-Judge systems to the optimization-based prompt injection attack.

☆ Enhanced Short Text Modeling: Leveraging Large Language Models for Topic Refinement

Shuyu Chang, Rui Wang, Peng Ren, Haiping Huang

Crafting effective topic models for brief texts, like tweets and news headlines, is essential for capturing the swift shifts in social dynamics. Traditional topic models, however, often fall short in accurately representing the semantic intricacies of short texts due to their brevity and lack of contextual data. In our study, we harness the advanced capabilities of Large Language Models (LLMs) to introduce a novel approach termed "Topic Refinement". This approach does not directly involve itself in the initial modeling of topics but focuses on improving topics after they have been mined. By employing prompt engineering, we direct LLMs to eliminate off-topic words within a given topic, ensuring that only contextually relevant words are preserved or substituted with ones that fit better semantically. This method emulates human-like scrutiny and improvement of topics, thereby elevating the semantic quality of the topics generated by various models. Our comprehensive evaluation across three unique datasets has shown that our topic refinement approach significantly enhances the semantic coherence of topics.

comment: 6 pages, 4 figures

☆ MEP: Multiple Kernel Learning Enhancing Relative Positional Encoding Length Extrapolation

Weiguo Gao

When the predicted sequence length exceeds the length seen during training, the transformer's inference accuracy diminishes. Existing relative position encoding methods, such as those based on the ALiBi technique, address the length extrapolation challenge exclusively through the implementation of a single kernel function, which introduces a constant bias to every post-softmax attention scores according to their distance. These approaches do not investigate or employ multiple kernel functions to address the extrapolation challenge. Drawing on the ALiBi approach, this study proposes a novel relative positional encoding method, called MEP, which employs a weighted average to combine distinct kernel functions(such as the exponential kernel and the Gaussian kernel) to generate a bias that is applied to post-softmax attention scores. Initially, the framework utilizes various kernel functions to construct multiple kernel functions. Each kernel function adheres to a consistent mean weight coefficient, harnessing the synergistic advantages of different kernels to formulate an innovative bias function. Subsequently, specific slopes are tailored for each kernel function, applying penalties at varying rates, to enhance the model's extrapolation capabilities. Finally, this bias is seamlessly incorporated as a penalty to the post-softmax scores. We present two distinct versions of our method: a parameter-free variant that requires no new learnable parameters, which enhances length extrapolation capabilities without compromising training efficiency, and a parameterized variant capable of integrating state-of-the-art techniques. Empirical evaluations across diverse datasets have demonstrated that both variants of our method achieve state-of-the-art performance, outperforming traditional parameter-free and parameterized approaches.

☆ ExpressEdit: Video Editing with Natural Language and Sketching

Bekzat Tilekbay, Saelyne Yang, Michal Lewkowicz, Alex Suryapranata, Juho Kim

Informational videos serve as a crucial source for explaining conceptual and procedural knowledge to novices and experts alike. When producing informational videos, editors edit videos by overlaying text/images or trimming footage to enhance the video quality and make it more engaging. However, video editing can be difficult and time-consuming, especially for novice video editors who often struggle with expressing and implementing their editing ideas. To address this challenge, we first explored how multimodality$-$natural language (NL) and sketching, which are natural modalities humans use for expression$-$can be utilized to support video editors in expressing video editing ideas. We gathered 176 multimodal expressions of editing commands from 10 video editors, which revealed the patterns of use of NL and sketching in describing edit intents. Based on the findings, we present ExpressEdit, a system that enables editing videos via NL text and sketching on the video frame. Powered by LLM and vision models, the system interprets (1) temporal, (2) spatial, and (3) operational references in an NL command and spatial references from sketching. The system implements the interpreted edits, which then the user can iterate on. An observational study (N=10) showed that ExpressEdit enhanced the ability of novice video editors to express and implement their edit ideas. The system allowed participants to perform edits more efficiently and generate more ideas by generating edits based on user's multimodal edit commands and supporting iterations on the editing commands. This work offers insights into the design of future multimodal interfaces and AI-based pipelines for video editing.

comment: 22 pages, 5 figures, to be published in ACM IUI 2024

☆ Solution for Emotion Prediction Competition of Workshop on Emotionally and Culturally Intelligent AI

Shengdong Xu, Zhouyang Chi, Yang Yang

This report provide a detailed description of the method that we explored and proposed in the WECIA Emotion Prediction Competition (EPC), which predicts a person's emotion through an artistic work with a comment. The dataset of this competition is ArtELingo, designed to encourage work on diversity across languages and cultures. The dataset has two main challenges, namely modal imbalance problem and language-cultural differences problem. In order to address this issue, we propose a simple yet effective approach called single-multi modal with Emotion-Cultural specific prompt(ECSP), which focuses on using the single modal message to enhance the performance of multimodal models and a well-designed prompt to reduce cultural differences problem. To clarify, our approach contains two main blocks: (1)XLM-R\cite{conneau2019unsupervised} based unimodal model and X$^2$-VLM\cite{zeng2022x} based multimodal model (2) Emotion-Cultural specific prompt. Our approach ranked first in the final test with a score of 0.627.

☆ Onboard deep lossless and near-lossless predictive coding of hyperspectral images with line-based attention

Diego Valsesia, Tiziano Bianchi, Enrico Magli

Deep learning methods have traditionally been difficult to apply to compression of hyperspectral images onboard of spacecrafts, due to the large computational complexity needed to achieve adequate representational power, as well as the lack of suitable datasets for training and testing. In this paper, we depart from the traditional autoencoder approach and we design a predictive neural network, called LineRWKV, that works recursively line-by-line to limit memory consumption. In order to achieve that, we adopt a novel hybrid attentive-recursive operation that combines the representational advantages of Transformers with the linear complexity and recursive implementation of recurrent neural networks. The compression algorithm performs prediction of each pixel using LineRWKV, followed by entropy coding of the residual. Experiments on the HySpecNet-11k dataset and PRISMA images show that LineRWKV is the first deep-learning method to outperform CCSDS-123.0-B-2 at lossless and near-lossless compression. Promising throughput results are also evaluated on a 7W embedded system.

☆ Depending on yourself when you should: Mentoring LLM with RL agents to become the master in cybersecurity games

Yikuan Yan, Yaolun Zhang, Keman Huang

Integrating LLM and reinforcement learning (RL) agent effectively to achieve complementary performance is critical in high stake tasks like cybersecurity operations. In this study, we introduce SecurityBot, a LLM agent mentored by pre-trained RL agents, to support cybersecurity operations. In particularly, the LLM agent is supported with a profile module to generated behavior guidelines, a memory module to accumulate local experiences, a reflection module to re-evaluate choices, and an action module to reduce action space. Additionally, it adopts the collaboration mechanism to take suggestions from pre-trained RL agents, including a cursor for dynamic suggestion taken, an aggregator for multiple mentors' suggestions ranking and a caller for proactive suggestion asking. Building on the CybORG experiment framework, our experiences show that SecurityBot demonstrates significant performance improvement compared with LLM or RL standalone, achieving the complementary performance in the cybersecurity games.

comment: 10 pages

☆ Language Models for Text Classification: Is In-Context Learning Enough? LREC

Aleksandra Edwards, Jose Camacho-Collados

Recent foundational language models have shown state-of-the-art performance in many NLP tasks in zero- and few-shot settings. An advantage of these models over more standard approaches based on fine-tuning is the ability to understand instructions written in natural language (prompts), which helps them generalise better to different tasks and domains without the need for specific training data. This makes them suitable for addressing text classification problems for domains with limited amounts of annotated instances. However, existing research is limited in scale and lacks understanding of how text generation models combined with prompting techniques compare to more established methods for text classification such as fine-tuning masked language models. In this paper, we address this research gap by performing a large-scale evaluation study for 16 text classification datasets covering binary, multiclass, and multilabel problems. In particular, we compare zero- and few-shot approaches of large language models to fine-tuning smaller language models. We also analyse the results by prompt, classification type, domain, and number of labels. In general, the results show how fine-tuning smaller and more efficient language models can still outperform few-shot approaches of larger language models, which have room for improvement when it comes to text classification.

comment: Accepted at LREC-COLING 2024

☆ SGHormer: An Energy-Saving Graph Transformer Driven by Spikes

Huizhe Zhang, Jintang Li, Liang Chen, Zibin Zheng

Graph Transformers (GTs) with powerful representation learning ability make a huge success in wide range of graph tasks. However, the costs behind outstanding performances of GTs are higher energy consumption and computational overhead. The complex structure and quadratic complexity during attention calculation in vanilla transformer seriously hinder its scalability on the large-scale graph data. Though existing methods have made strides in simplifying combinations among blocks or attention-learning paradigm to improve GTs' efficiency, a series of energy-saving solutions originated from biologically plausible structures are rarely taken into consideration when constructing GT framework. To this end, we propose a new spiking-based graph transformer (SGHormer). It turns full-precision embeddings into sparse and binarized spikes to reduce memory and computational costs. The spiking graph self-attention and spiking rectify blocks in SGHormer explicitly capture global structure information and recover the expressive power of spiking embeddings, respectively. In experiments, SGHormer achieves comparable performances to other full-precision GTs with extremely low computational energy consumption. The results show that SGHomer makes a remarkable progress in the field of low-energy GTs.

comment: 9 pages, 3 figures

☆ An Extension-based Approach for Computing and Verifying Preferences in Abstract Argumentation

Quratul-ain Mahesar, Nir Oren, Wamberto W. Vasconcelos

We present an extension-based approach for computing and verifying preferences in an abstract argumentation system. Although numerous argumentation semantics have been developed previously for identifying acceptable sets of arguments from an argumentation framework, there is a lack of justification behind their acceptability based on implicit argument preferences. Preference-based argumentation frameworks allow one to determine what arguments are justified given a set of preferences. Our research considers the inverse of the standard reasoning problem, i.e., given an abstract argumentation framework and a set of justified arguments, we compute what the possible preferences over arguments are. Furthermore, there is a need to verify (i.e., assess) that the computed preferences would lead to the acceptable sets of arguments. This paper presents a novel approach and algorithm for exhaustively computing and enumerating all possible sets of preferences (restricted to three identified cases) for a conflict-free set of arguments in an abstract argumentation framework. We prove the soundness, completeness and termination of the algorithm. The research establishes that preferences are determined using an extension-based approach after the evaluation phase (acceptability of arguments) rather than stated beforehand. In this work, we focus our research study on grounded, preferred and stable semantics. We show that the complexity of computing sets of preferences is exponential in the number of arguments, and thus, describe an approximate approach and algorithm to compute the preferences. Furthermore, we present novel algorithms for verifying (i.e., assessing) the computed preferences. We provide details of the implementation of the algorithms (source code has been made available), various experiments performed to evaluate the algorithms and the analysis of the results.

☆ S+t-SNE - Bringing dimensionality reduction to data streams

Pedro C. Vieira, João P. Montrezol, João T. Vieira, João Gama

We present S+t-SNE, an adaptation of the t-SNE algorithm designed to handle infinite data streams. The core idea behind S+t-SNE is to update the t-SNE embedding incrementally as new data arrives, ensuring scalability and adaptability to handle streaming scenarios. By selecting the most important points at each step, the algorithm ensures scalability while keeping informative visualisations. Employing a blind method for drift management adjusts the embedding space, facilitating continuous visualisation of evolving data dynamics. Our experimental evaluations demonstrate the effectiveness and efficiency of S+t-SNE. The results highlight its ability to capture patterns in a streaming scenario. We hope our approach offers researchers and practitioners a real-time tool for understanding and interpreting high-dimensional data.

comment: This preprint has not undergone peer review or any post-submission improvements or corrections. We will soon add a link to the final version of this contribution that underwent peer-review and post-acceptance improvements and was presented at IDA2024 (https://ida2024.org/)

☆ PeersimGym: An Environment for Solving the Task Offloading Problem with Reinforcement Learning

Frederico Metelo, Stevo Racković, Pedro Ákos, Cláudia Soares

Task offloading, crucial for balancing computational loads across devices in networks such as the Internet of Things, poses significant optimization challenges, including minimizing latency and energy usage under strict communication and storage constraints. While traditional optimization falls short in scalability; and heuristic approaches lack in achieving optimal outcomes, Reinforcement Learning (RL) offers a promising avenue by enabling the learning of optimal offloading strategies through iterative interactions. However, the efficacy of RL hinges on access to rich datasets and custom-tailored, realistic training environments. To address this, we introduce PeersimGym, an open-source, customizable simulation environment tailored for developing and optimizing task offloading strategies within computational networks. PeersimGym supports a wide range of network topologies and computational constraints and integrates a \textit{PettingZoo}-based interface for RL agent deployment in both solo and multi-agent setups. Furthermore, we demonstrate the utility of the environment through experiments with Deep Reinforcement Learning agents, showcasing the potential of RL-based approaches to significantly enhance offloading strategies in distributed computing settings. PeersimGym thus bridges the gap between theoretical RL models and their practical applications, paving the way for advancements in efficient task offloading methodologies.

☆ UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection with Sparse LiDAR and Large Domain Gaps

Maciej K Wozniak, Mattias Hansson, Marko Thiel, Patric Jensfelt

In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains. Our code is open-source and will be available soon.

★ Data-driven Energy Consumption Modelling for Electric Micromobility using an Open Dataset

Yue Ding, Sen Yan, Maqsood Hussain Shah, Hongyuan Fang, Ji Li, Mingming Liu

The escalating challenges of traffic congestion and environmental degradation underscore the critical importance of embracing E-Mobility solutions in urban spaces. In particular, micro E-Mobility tools such as E-scooters and E-bikes, play a pivotal role in this transition, offering sustainable alternatives for urban commuters. However, the energy consumption patterns for these tools are a critical aspect that impacts their effectiveness in real-world scenarios and is essential for trip planning and boosting user confidence in using these. To this effect, recent studies have utilised physical models customised for specific mobility tools and conditions, but these models struggle with generalization and effectiveness in real-world scenarios due to a notable absence of open datasets for thorough model evaluation and verification. To fill this gap, our work presents an open dataset, collected in Dublin, Ireland, specifically designed for energy modelling research related to E-Scooters and E-Bikes. Furthermore, we provide a comprehensive analysis of energy consumption modelling based on the dataset using a set of representative machine learning algorithms and compare their performance against the contemporary mathematical models as a baseline. Our results demonstrate a notable advantage for data-driven models in comparison to the corresponding mathematical models for estimating energy consumption. Specifically, data-driven models outperform physical models in accuracy by up to 83.83% for E-Bikes and 82.16% for E-Scooters based on an in-depth analysis of the dataset under certain assumptions.

comment: 7 pages, 5 figures, 4 tables. This manuscript has been accepted by the IEEE ITEC 2024

☆ Denoising Table-Text Retrieval for Open-Domain Question Answering LREC

Deokhyung Kang, Baikjin Jung, Yunsu Kim, Gary Geunbae Lee

In table-text open-domain question answering, a retriever system retrieves relevant evidence from tables and text to answer questions. Previous studies in table-text open-domain question answering have two common challenges: firstly, their retrievers can be affected by false-positive labels in training datasets; secondly, they may struggle to provide appropriate evidence for questions that require reasoning across the table. To address these issues, we propose Denoised Table-Text Retriever (DoTTeR). Our approach involves utilizing a denoised training dataset with fewer false positive labels by discarding instances with lower question-relevance scores measured through a false positive detection model. Subsequently, we integrate table-level ranking information into the retriever to assist in finding evidence for questions that demand reasoning across the table. To encode this ranking information, we fine-tune a rank-aware column encoder to identify minimum and maximum values within a column. Experimental results demonstrate that DoTTeR significantly outperforms strong baselines on both retrieval recall and downstream QA tasks. Our code is available at https://github.com/deokhk/DoTTeR.

comment: Accepted to LREC-COLING 2024

☆ Fully-fused Multi-Layer Perceptrons on Intel Data Center GPUs

Kai Yuan, Christoph Bauinger, Xiangyi Zhang, Pascal Baehr, Matthias Kirchhart, Darius Dabert, Adrien Tousnakhoff, Pierre Boudier, Michael Paulitsch

This paper presents a SYCL implementation of Multi-Layer Perceptrons (MLPs), which targets and is optimized for the Intel Data Center GPU Max 1550. To increase the performance, our implementation minimizes the slow global memory accesses by maximizing the data reuse within the general register file and the shared local memory by fusing the operations in each layer of the MLP. We show with a simple roofline model that this results in a significant increase in the arithmetic intensity, leading to improved performance, especially for inference. We compare our approach to a similar CUDA implementation for MLPs and show that our implementation on the Intel Data Center GPU outperforms the CUDA implementation on Nvidia's H100 GPU by a factor up to 2.84 in inference and 1.75 in training. The paper also showcases the efficiency of our SYCL implementation in three significant areas: Image Compression, Neural Radiance Fields, and Physics-Informed Machine Learning. In all cases, our implementation outperforms the off-the-shelf Intel Extension for PyTorch (IPEX) implementation on the same Intel GPU by up to a factor of 30 and the CUDA PyTorch version on Nvidia's H100 GPU by up to a factor 19. The code can be found at https://github.com/intel/tiny-dpcpp-nn.

☆ LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation

Ke Guo, Zhenwei Miao, Wei Jing, Weiwei Liu, Weizi Li, Dayang Hao, Jia Pan

Microscopic traffic simulation plays a crucial role in transportation engineering by providing insights into individual vehicle behavior and overall traffic flow. However, creating a realistic simulator that accurately replicates human driving behaviors in various traffic conditions presents significant challenges. Traditional simulators relying on heuristic models often fail to deliver accurate simulations due to the complexity of real-world traffic environments. Due to the covariate shift issue, existing imitation learning-based simulators often fail to generate stable long-term simulations. In this paper, we propose a novel approach called learner-aware supervised imitation learning to address the covariate shift problem in multi-agent imitation learning. By leveraging a variational autoencoder simultaneously modeling the expert and learner state distribution, our approach augments expert states such that the augmented state is aware of learner state distribution. Our method, applied to urban traffic simulation, demonstrates significant improvements over existing state-of-the-art baselines in both short-term microscopic and long-term macroscopic realism when evaluated on the real-world dataset pNEUMA.

comment: accepted by cvpr 2024. arXiv admin note: text overlap with arXiv:2306.06401

☆ Dual Memory Networks: A Versatile Adaptation Approach for Vision-Language Models CVPR2024

Yabin Zhang, Wenjie Zhu, Hui Tang, Zhiyuan Ma, Kaiyang Zhou, Lei Zhang

With the emergence of pre-trained vision-language models like CLIP, how to adapt them to various downstream classification tasks has garnered significant attention in recent research. The adaptation strategies can be typically categorized into three paradigms: zero-shot adaptation, few-shot adaptation, and the recently-proposed training-free few-shot adaptation. Most existing approaches are tailored for a specific setting and can only cater to one or two of these paradigms. In this paper, we introduce a versatile adaptation approach that can effectively work under all three settings. Specifically, we propose the dual memory networks that comprise dynamic and static memory components. The static memory caches training data knowledge, enabling training-free few-shot adaptation, while the dynamic memory preserves historical test features online during the testing process, allowing for the exploration of additional data insights beyond the training set. This novel capability enhances model performance in the few-shot setting and enables model usability in the absence of training data. The two memory networks employ the same flexible memory interactive strategy, which can operate in a training-free mode and can be further enhanced by incorporating learnable projection layers. Our approach is tested across 11 datasets under the three task settings. Remarkably, in the zero-shot scenario, it outperforms existing methods by over 3\% and even shows superior results against methods utilizing external training data. Additionally, our method exhibits robust performance against natural distribution shifts. Codes are available at \url{https://github.com/YBZh/DMN}.

comment: CVPR2024; Codes are available at \url{https://github.com/YBZh/DMN}

☆ Parameterized Analysis of Bribery in Challenge the Champ Tournaments

Juhi Chaudhary, Hendrik Molter, Meirav Zehavi

Challenge the champ tournaments are one of the simplest forms of competition, where a (initially selected) champ is repeatedly challenged by other players. If a player beats the champ, then that player is considered the new (current) champ. Each player in the competition challenges the current champ once in a fixed order. The champ of the last round is considered the winner of the tournament. We investigate a setting where players can be bribed to lower their winning probability against the initial champ. The goal is to maximize the probability of the initial champ winning the tournament by bribing the other players, while not exceeding a given budget for the bribes. Mattei et al. [Journal of Applied Logic, 2015] showed that the problem can be solved in pseudo-polynomial time, and that it is in XP when parameterized by the number of players. We show that the problem is weakly NP-hard and W[1]-hard when parameterized by the number of players. On the algorithmic side, we show that the problem is fixed-parameter tractable when parameterized either by the number of different bribe values or the number of different probability values. To this end, we establish several results that are of independent interest. In particular, we show that the product knapsack problem is W[1]-hard when parameterized by the number of items in the knapsack, and that constructive bribery for cup tournaments is W[1]-hard when parameterized by the number of players. Furthermore, we present a novel way of designing mixed integer linear programs, ensuring optimal solutions where all variables are integers.

☆ Towards a Zero-Data, Controllable, Adaptive Dialog System

Dirk Väth, Lindsey Vanderlyn, Ngoc Thang Vu

Conversational Tree Search (V\"ath et al., 2023) is a recent approach to controllable dialog systems, where domain experts shape the behavior of a Reinforcement Learning agent through a dialog tree. The agent learns to efficiently navigate this tree, while adapting to information needs, e.g., domain familiarity, of different users. However, the need for additional training data hinders deployment in new domains. To address this, we explore approaches to generate this data directly from dialog trees. We improve the original approach, and show that agents trained on synthetic data can achieve comparable dialog success to models trained on human data, both when using a commercial Large Language Model for generation, or when using a smaller open-source model, running on a single GPU. We further demonstrate the scalability of our approach by collecting and testing on two new datasets: ONBOARD, a new domain helping foreign residents moving to a new city, and the medical domain DIAGNOSE, a subset of Wikipedia articles related to scalp and head symptoms. Finally, we perform human testing, where no statistically significant differences were found in either objective or subjective measures between models trained on human and generated data.

☆ m3P: Towards Multimodal Multilingual Translation with Multimodal Prompt COLING 2024

Jian Yang, Hongcheng Guo, Yuwei Yin, Jiaqi Bai, Bing Wang, Jiaheng Liu, Xinnian Liang, Linzheng Cahi, Liqun Yang, Zhoujun Li

Multilingual translation supports multiple translation directions by projecting all languages in a shared space, but the translation quality is undermined by the difference between languages in the text-only modality, especially when the number of languages is large. To bridge this gap, we introduce visual context as the universal language-independent representation to facilitate multilingual translation. In this paper, we propose a framework to leverage the multimodal prompt to guide the Multimodal Multilingual neural Machine Translation (m3P), which aligns the representations of different languages with the same meaning and generates the conditional vision-language memory for translation. We construct a multilingual multimodal instruction dataset (InstrMulti102) to support 102 languages. Our method aims to minimize the representation distance of different languages by regarding the image as a central language. Experimental results show that m3P outperforms previous text-only baselines and multilingual multimodal methods by a large margin. Furthermore, the probing experiments validate the effectiveness of our method in enhancing translation under the low-resource and massively multilingual scenario.

comment: COLING 2024

☆ Practical Applications of Advanced Cloud Services and Generative AI Systems in Medical Image Analysis

Jingyu Xu, Binbin Wu, Jiaxin Huang, Yulu Gong, Yifan Zhang, Bo Liu

The medical field is one of the important fields in the application of artificial intelligence technology. With the explosive growth and diversification of medical data, as well as the continuous improvement of medical needs and challenges, artificial intelligence technology is playing an increasingly important role in the medical field. Artificial intelligence technologies represented by computer vision, natural language processing, and machine learning have been widely penetrated into diverse scenarios such as medical imaging, health management, medical information, and drug research and development, and have become an important driving force for improving the level and quality of medical services.The article explores the transformative potential of generative AI in medical imaging, emphasizing its ability to generate syntheticACM-2 data, enhance images, aid in anomaly detection, and facilitate image-to-image translation. Despite challenges like model complexity, the applications of generative models in healthcare, including Med-PaLM 2 technology, show promising results. By addressing limitations in dataset size and diversity, these models contribute to more accurate diagnoses and improved patient outcomes. However, ethical considerations and collaboration among stakeholders are essential for responsible implementation. Through experiments leveraging GANs to augment brain tumor MRI datasets, the study demonstrates how generative AI can enhance image quality and diversity, ultimately advancing medical diagnostics and patient care.

☆ VDSC: Enhancing Exploration Timing with Value Discrepancy and State Counts

Marius Captari, Remo Sasso, Matthia Sabatelli

Despite the considerable attention given to the questions of \textit{how much} and \textit{how to} explore in deep reinforcement learning, the investigation into \textit{when} to explore remains relatively less researched. While more sophisticated exploration strategies can excel in specific, often sparse reward environments, existing simpler approaches, such as $\epsilon$-greedy, persist in outperforming them across a broader spectrum of domains. The appeal of these simpler strategies lies in their ease of implementation and generality across a wide range of domains. The downside is that these methods are essentially a blind switching mechanism, which completely disregards the agent's internal state. In this paper, we propose to leverage the agent's internal state to decide \textit{when} to explore, addressing the shortcomings of blind switching mechanisms. We present Value Discrepancy and State Counts through homeostasis (VDSC), a novel approach for efficient exploration timing. Experimental results on the Atari suite demonstrate the superiority of our strategy over traditional methods such as $\epsilon$-greedy and Boltzmann, as well as more sophisticated techniques like Noisy Nets.

☆ KC-GenRe: A Knowledge-constrained Generative Re-ranking Method Based on Large Language Models for Knowledge Graph Completion LREC

Yilin Wang, Minghao Hu, Zhen Huang, Dongsheng Li, Dong Yang, Xicheng Lu

The goal of knowledge graph completion (KGC) is to predict missing facts among entities. Previous methods for KGC re-ranking are mostly built on non-generative language models to obtain the probability of each candidate. Recently, generative large language models (LLMs) have shown outstanding performance on several tasks such as information extraction and dialog systems. Leveraging them for KGC re-ranking is beneficial for leveraging the extensive pre-trained knowledge and powerful generative capabilities. However, it may encounter new problems when accomplishing the task, namely mismatch, misordering and omission. To this end, we introduce KC-GenRe, a knowledge-constrained generative re-ranking method based on LLMs for KGC. To overcome the mismatch issue, we formulate the KGC re-ranking task as a candidate identifier sorting generation problem implemented by generative LLMs. To tackle the misordering issue, we develop a knowledge-guided interactive training method that enhances the identification and ranking of candidates. To address the omission issue, we design a knowledge-augmented constrained inference method that enables contextual prompting and controlled generation, so as to obtain valid rankings. Experimental results show that KG-GenRe achieves state-of-the-art performance on four datasets, with gains of up to 6.7% and 7.7% in the MRR and Hits@1 metric compared to previous methods, and 9.0% and 11.1% compared to that without re-ranking. Extensive analysis demonstrates the effectiveness of components in KG-GenRe.

comment: This paper has been accepted for publication in the proceedings of LREC-COLING 2024

☆ Boosting Few-Shot Learning with Disentangled Self-Supervised Learning and Meta-Learning for Medical Image Classification

Eva Pachetti, Sotirios A. Tsaftaris, Sara Colantonio

Background and objective: Employing deep learning models in critical domains such as medical imaging poses challenges associated with the limited availability of training data. We present a strategy for improving the performance and generalization capabilities of models trained in low-data regimes. Methods: The proposed method starts with a pre-training phase, where features learned in a self-supervised learning setting are disentangled to improve the robustness of the representations for downstream tasks. We then introduce a meta-fine-tuning step, leveraging related classes between meta-training and meta-testing phases but varying the granularity level. This approach aims to enhance the model's generalization capabilities by exposing it to more challenging classification tasks during meta-training and evaluating it on easier tasks but holding greater clinical relevance during meta-testing. We demonstrate the effectiveness of the proposed approach through a series of experiments exploring several backbones, as well as diverse pre-training and fine-tuning schemes, on two distinct medical tasks, i.e., classification of prostate cancer aggressiveness from MRI data and classification of breast cancer malignity from microscopic images. Results: Our results indicate that the proposed approach consistently yields superior performance w.r.t. ablation experiments, maintaining competitiveness even when a distribution shift between training and evaluation data occurs. Conclusion: Extensive experiments demonstrate the effectiveness and wide applicability of the proposed approach. We hope that this work will add another solution to the arsenal of addressing learning issues in data-scarce imaging domains.

comment: 20 pages, 4 figures, 4 tables. Submitted to Elsevier on 25 March 2024

☆ Equipping Sketch Patches with Context-Aware Positional Encoding for Graphic Sketch Representation

Sicong Zang, Zhijun Fang

The drawing order of a sketch records how it is created stroke-by-stroke by a human being. For graphic sketch representation learning, recent studies have injected sketch drawing orders into graph edge construction by linking each patch to another in accordance to a temporal-based nearest neighboring strategy. However, such constructed graph edges may be unreliable, since a sketch could have variants of drawings. In this paper, we propose a variant-drawing-protected method by equipping sketch patches with context-aware positional encoding (PE) to make better use of drawing orders for learning graphic sketch representation. Instead of injecting sketch drawings into graph edges, we embed these sequential information into graph nodes only. More specifically, each patch embedding is equipped with a sinusoidal absolute PE to highlight the sequential position in the drawing order. And its neighboring patches, ranked by the values of self-attention scores between patch embeddings, are equipped with learnable relative PEs to restore the contextual positions within a neighborhood. During message aggregation via graph convolutional networks, a node receives both semantic contents from patch embeddings and contextual patterns from PEs by its neighbors, arriving at drawing-order-enhanced sketch representations. Experimental results indicate that our method significantly improves sketch healing and controllable sketch synthesis.

☆ MapGuide: A Simple yet Effective Method to Reconstruct Continuous Language from Brain Activities NAACL 2024

Xinpei Zhao, Jingyuan Sun, Shaonan Wang, Jing Ye, Xiaohan Zhang, Chengqing Zong

Decoding continuous language from brain activity is a formidable yet promising field of research. It is particularly significant for aiding people with speech disabilities to communicate through brain signals. This field addresses the complex task of mapping brain signals to text. The previous best attempt reverse-engineered this process in an indirect way: it began by learning to encode brain activity from text and then guided text generation by aligning with predicted brain responses. In contrast, we propose a simple yet effective method that guides text reconstruction by directly comparing them with the predicted text embeddings mapped from brain activities. Comprehensive experiments reveal that our method significantly outperforms the current state-of-the-art model, showing average improvements of 77% and 54% on BLEU and METEOR scores. We further validate the proposed modules through detailed ablation studies and case analyses and highlight a critical correlation: the more precisely we map brain activities to text embeddings, the better the text reconstruction results. Such insight can simplify the task of reconstructing language from brain activities for future work, emphasizing the importance of improving brain-to-text-embedding mapping techniques.

comment: Accepted to NAACL 2024 main conference

☆ Prediction-sharing During Training and Inference

Yotam Gafni, Ronen Gradwohl, Moshe Tennenholtz

Two firms are engaged in a competitive prediction task. Each firm has two sources of data -- labeled historical data and unlabeled inference-time data -- and uses the former to derive a prediction model, and the latter to make predictions on new instances. We study data-sharing contracts between the firms. The novelty of our study is to introduce and highlight the differences between contracts that share prediction models only, contracts to share inference-time predictions only, and contracts to share both. Our analysis proceeds on three levels. First, we develop a general Bayesian framework that facilitates our study. Second, we narrow our focus to two natural settings within this framework: (i) a setting in which the accuracy of each firm's prediction model is common knowledge, but the correlation between the respective models is unknown; and (ii) a setting in which two hypotheses exist regarding the optimal predictor, and one of the firms has a structural advantage in deducing it. Within these two settings we study optimal contract choice. More specifically, we find the individually rational and Pareto-optimal contracts for some notable cases, and describe specific settings where each of the different sharing contracts emerge as optimal. Finally, in the third level of our analysis we demonstrate the applicability of our concepts in a synthetic simulation using real loan data.

☆ Natural Language Requirements Testability Measurement Based on Requirement Smells

Morteza Zakeri-Nasrabadi, Saeed Parsa

Requirements form the basis for defining software systems' obligations and tasks. Testable requirements help prevent failures, reduce maintenance costs, and make it easier to perform acceptance tests. However, despite the importance of measuring and quantifying requirements testability, no automatic approach for measuring requirements testability has been proposed based on the requirements smells, which are at odds with the requirements testability. This paper presents a mathematical model to evaluate and rank the natural language requirements testability based on an extensive set of nine requirements smells, detected automatically, and acceptance test efforts determined by requirement length and its application domain. Most of the smells stem from uncountable adjectives, context-sensitive, and ambiguous words. A comprehensive dictionary is required to detect such words. We offer a neural word-embedding technique to generate such a dictionary automatically. Using the dictionary, we could automatically detect Polysemy smell (domain-specific ambiguity) for the first time in 10 application domains. Our empirical study on nearly 1000 software requirements from six well-known industrial and academic projects demonstrates that the proposed smell detection approach outperforms Smella, a state-of-the-art tool, in detecting requirements smells. The precision and recall of smell detection are improved with an average of 0.03 and 0.33, respectively, compared to the state-of-the-art. The proposed requirement testability model measures the testability of 985 requirements with a mean absolute error of 0.12 and a mean squared error of 0.03, demonstrating the model's potential for practical use.

comment: 45 pages, 16 figures, and 13 tables; submitted as a journal paper

☆ A Unified Kernel for Neural Network Learning

Shao-Qun Zhang, Zong-Yi Chen, Yong-Ming Tian, Xun Lu

Past decades have witnessed a great interest in the distinction and connection between neural network learning and kernel learning. Recent advancements have made theoretical progress in connecting infinite-wide neural networks and Gaussian processes. Two predominant approaches have emerged: the Neural Network Gaussian Process (NNGP) and the Neural Tangent Kernel (NTK). The former, rooted in Bayesian inference, represents a zero-order kernel, while the latter, grounded in the tangent space of gradient descents, is a first-order kernel. In this paper, we present the Unified Neural Kernel (UNK), which characterizes the learning dynamics of neural networks with gradient descents and parameter initialization. The proposed UNK kernel maintains the limiting properties of both NNGP and NTK, exhibiting behaviors akin to NTK with a finite learning step and converging to NNGP as the learning step approaches infinity. Besides, we also theoretically characterize the uniform tightness and learning convergence of the UNK kernel, providing comprehensive insights into this unified kernel. Experimental results underscore the effectiveness of our proposed method.

☆ LaRE^2: Latent Reconstruction Error Based Method for Diffusion-Generated Image Detection CVPR 2024

Yunpeng Luo, Junlong Du, Ke Yan, Shouhong Ding

The evolution of Diffusion Models has dramatically improved image generation quality, making it increasingly difficult to differentiate between real and generated images. This development, while impressive, also raises significant privacy and security concerns. In response to this, we propose a novel Latent REconstruction error guided feature REfinement method (LaRE^2) for detecting the diffusion-generated images. We come up with the Latent Reconstruction Error (LaRE), the first reconstruction-error based feature in the latent space for generated image detection. LaRE surpasses existing methods in terms of feature extraction efficiency while preserving crucial cues required to differentiate between the real and the fake. To exploit LaRE, we propose an Error-Guided feature REfinement module (EGRE), which can refine the image feature guided by LaRE to enhance the discriminativeness of the feature. Our EGRE utilizes an align-then-refine mechanism, which effectively refines the image feature for generated-image detection from both spatial and channel perspectives. Extensive experiments on the large-scale GenImage benchmark demonstrate the superiority of our LaRE^2, which surpasses the best SoTA method by up to 11.9%/12.1% average ACC/AP across 8 different image generators. LaRE also surpasses existing methods in terms of feature extraction cost, delivering an impressive speed enhancement of 8 times.

comment: CVPR 2024

☆ Incorporating Exponential Smoothing into MLP: A Simple but Effective Sequence Model

Jiqun Chu, Zuoquan Lin

Modeling long-range dependencies in sequential data is a crucial step in sequence learning. A recently developed model, the Structured State Space (S4), demonstrated significant effectiveness in modeling long-range sequences. However, It is unclear whether the success of S4 can be attributed to its intricate parameterization and HiPPO initialization or simply due to State Space Models (SSMs). To further investigate the potential of the deep SSMs, we start with exponential smoothing (ETS), a simple SSM, and propose a stacked architecture by directly incorporating it into an element-wise MLP. We augment simple ETS with additional parameters and complex field to reduce the inductive bias. Despite increasing less than 1\% of parameters of element-wise MLP, our models achieve comparable results to S4 on the LRA benchmark.

comment: 12 pages, 5 tables, 3 figures

☆ Aligning Large Language Models for Enhancing Psychiatric Interviews through Symptom Delineation and Summarization

Jae-hee So, Joonhwan Chang, Eunji Kim, Junho Na, JiYeon Choi, Jy-yong Sohn, Byung-Hoon Kim, Sang Hui Chu

Recent advancements in Large Language Models (LLMs) have accelerated their usage in various domains. Given the fact that psychiatric interviews are goal-oriented and structured dialogues between the professional interviewer and the interviewee, it is one of the most underexplored areas where LLMs can contribute substantial value. Here, we explore the use of LLMs for enhancing psychiatric interviews, by analyzing counseling data from North Korean defectors with traumatic events and mental health issues. Specifically, we investigate whether LLMs can (1) delineate the part of the conversation that suggests psychiatric symptoms and name the symptoms, and (2) summarize stressors and symptoms, based on the interview dialogue transcript. Here, the transcript data was labeled by mental health experts for training and evaluation of LLMs. Our experimental results show that appropriately prompted LLMs can achieve high performance on both the symptom delineation task and the summarization task. This research contributes to the nascent field of applying LLMs to psychiatric interview and demonstrates their potential effectiveness in aiding mental health practitioners.

☆ Knowledge-Powered Recommendation for an Improved Diet Water Footprint AAAI'24

Saurav Joshi, Filip Ilievski, Jay Pujara

According to WWF, 1.1 billion people lack access to water, and 2.7 billion experience water scarcity at least one month a year. By 2025, two-thirds of the world's population may be facing water shortages. This highlights the urgency of managing water usage efficiently, especially in water-intensive sectors like food. This paper proposes a recommendation engine, powered by knowledge graphs, aiming to facilitate sustainable and healthy food consumption. The engine recommends ingredient substitutes in user recipes that improve nutritional value and reduce environmental impact, particularly water footprint. The system architecture includes source identification, information extraction, schema alignment, knowledge graph construction, and user interface development. The research offers a promising tool for promoting healthier eating habits and contributing to water conservation efforts.

comment: 3 pages, 1 figure, AAAI'24

☆ AI Safety: Necessary, but insufficient and possibly problematic

Deepak P

This article critically examines the recent hype around AI safety. We first start with noting the nature of the AI safety hype as being dominated by governments and corporations, and contrast it with other avenues within AI research on advancing social good. We consider what 'AI safety' actually means, and outline the dominant concepts that the digital footprint of AI safety aligns with. We posit that AI safety has a nuanced and uneasy relationship with transparency and other allied notions associated with societal good, indicating that it is an insufficient notion if the goal is that of societal good in a broad sense. We note that the AI safety debate has already influenced some regulatory efforts in AI, perhaps in not so desirable directions. We also share our concerns on how AI safety may normalize AI that advances structural harm through providing exploitative and harmful AI with a veneer of safety.

☆ On permutation-invariant neural networks

Masanari Kimura, Ryotaro Shimizu, Yuki Hirakawa, Ryosuke Goto, Yuki Saito

Conventional machine learning algorithms have traditionally been designed under the assumption that input data follows a vector-based format, with an emphasis on vector-centric paradigms. However, as the demand for tasks involving set-based inputs has grown, there has been a paradigm shift in the research community towards addressing these challenges. In recent years, the emergence of neural network architectures such as Deep Sets and Transformers has presented a significant advancement in the treatment of set-based data. These architectures are specifically engineered to naturally accommodate sets as input, enabling more effective representation and processing of set structures. Consequently, there has been a surge of research endeavors dedicated to exploring and harnessing the capabilities of these architectures for various tasks involving the approximation of set functions. This comprehensive survey aims to provide an overview of the diverse problem settings and ongoing research efforts pertaining to neural networks that approximate set functions. By delving into the intricacies of these approaches and elucidating the associated challenges, the survey aims to equip readers with a comprehensive understanding of the field. Through this comprehensive perspective, we hope that researchers can gain valuable insights into the potential applications, inherent limitations, and future directions of set-based neural networks. Indeed, from this survey we gain two insights: i) Deep Sets and its variants can be generalized by differences in the aggregation function, and ii) the behavior of Deep Sets is sensitive to the choice of the aggregation function. From these observations, we show that Deep Sets, one of the well-known permutation-invariant neural networks, can be generalized in the sense of a quasi-arithmetic mean.

☆ Transcribing Bengali Text with Regional Dialects to IPA using District Guided Tokens

S M Jishanul Islam, Sadia Ahmmed, Sahid Hossain Mustakim

Accurate transcription of Bengali text to the International Phonetic Alphabet (IPA) is a challenging task due to the complex phonology of the language and context-dependent sound changes. This challenge is even more for regional Bengali dialects due to unavailability of standardized spelling conventions for these dialects, presence of local and foreign words popular in those regions and phonological diversity across different regions. This paper presents an approach to this sequence-to-sequence problem by introducing the District Guided Tokens (DGT) technique on a new dataset spanning six districts of Bangladesh. The key idea is to provide the model with explicit information about the regional dialect or "district" of the input text before generating the IPA transcription. This is achieved by prepending a district token to the input sequence, effectively guiding the model to understand the unique phonetic patterns associated with each district. The DGT technique is applied to fine-tune several transformer-based models, on this new dataset. Experimental results demonstrate the effectiveness of DGT, with the ByT5 model achieving superior performance over word-based models like mT5, BanglaT5, and umT5. This is attributed to ByT5's ability to handle a high percentage of out-of-vocabulary words in the test set. The proposed approach highlights the importance of incorporating regional dialect information into ubiquitous natural language processing systems for languages with diverse phonological variations. The following work was a result of the "Bhashamul" challenge, which is dedicated to solving the problem of Bengali text with regional dialects to IPA transcription https://www.kaggle.com/competitions/regipa/. The training and inference notebooks are available through the competition link.

comment: This work became the champion of the Bhashamul challenge

☆ An Open-source End-to-End Logic Optimization Framework for Large-scale Boolean Network with Reinforcement Learning

Zhen Li, Kaixiang Zhu, Xuegong Zhou, Lingli Wang

We propose an open-source end-to-end logic optimization framework for large-scale boolean network with reinforcement learning.

comment: 5 pages, 4 figures, 1 table

☆ ELLEN: Extremely Lightly Supervised Learning For Efficient Named Entity Recognition LREC

Haris Riaz, Razvan-Gabriel Dumitru, Mihai Surdeanu

In this work, we revisit the problem of semi-supervised named entity recognition (NER) focusing on extremely light supervision, consisting of a lexicon containing only 10 examples per class. We introduce ELLEN, a simple, fully modular, neuro-symbolic method that blends fine-tuned language models with linguistic rules. These rules include insights such as ''One Sense Per Discourse'', using a Masked Language Model as an unsupervised NER, leveraging part-of-speech tags to identify and eliminate unlabeled entities as false negatives, and other intuitions about classifier confidence scores in local and global context. ELLEN achieves very strong performance on the CoNLL-2003 dataset when using the minimal supervision from the lexicon above. It also outperforms most existing (and considerably more complex) semi-supervised NER methods under the same supervision settings commonly used in the literature (i.e., 5% of the training data). Further, we evaluate our CoNLL-2003 model in a zero-shot scenario on WNUT-17 where we find that it outperforms GPT-3.5 and achieves comparable performance to GPT-4. In a zero-shot setting, ELLEN also achieves over 75% of the performance of a strong, fully supervised model trained on gold data. Our code is available at: https://github.com/hriaz17/ELLEN.

comment: Accepted to LREC-COLING 2024

☆ Explainable Graph Neural Networks for Observation Impact Analysis in Atmospheric State Estimation

Hyeon-Ju Jeon, Jeon-Ho Kang, In-Hyuk Kwon, O-Joun Lee

This paper investigates the impact of observations on atmospheric state estimation in weather forecasting systems using graph neural networks (GNNs) and explainability methods. We integrate observation and Numerical Weather Prediction (NWP) points into a meteorological graph, extracting $k$-hop subgraphs centered on NWP points. Self-supervised GNNs are employed to estimate the atmospheric state by aggregating data within these $k$-hop radii. The study applies gradient-based explainability methods to quantify the significance of different observations in the estimation process. Evaluated with data from 11 satellite and land-based observations, the results highlight the effectiveness of visualizing the importance of observation types, enhancing the understanding and optimization of observational data in weather forecasting.

☆ Application-Driven Innovation in Machine Learning

David Rolnick, Alan Aspuru-Guzik, Sara Beery, Bistra Dilkina, Priya L. Donti, Marzyeh Ghassemi, Hannah Kerner, Claire Monteleoni, Esther Rolf, Milind Tambe, Adam White

As applications of machine learning proliferate, innovative algorithms inspired by specific real-world challenges have become increasingly important. Such work offers the potential for significant impact not merely in domains of application but also in machine learning itself. In this paper, we describe the paradigm of application-driven research in machine learning, contrasting it with the more standard paradigm of methods-driven research. We illustrate the benefits of application-driven machine learning and how this approach can productively synergize with methods-driven work. Despite these benefits, we find that reviewing, hiring, and teaching practices in machine learning often hold back application-driven innovation. We outline how these processes may be improved.

comment: 12 pages, 3 figures

☆ Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance

Donghoon Ahn, Hyoungwon Cho, Jaewon Min, Wooseok Jang, Jungwoo Kim, SeonHwa Kim, Hyun Hee Park, Kyong Hwan Jin, Seungryong Kim

Recent studies have demonstrated that diffusion models are capable of generating high-quality samples, but their quality heavily depends on sampling guidance techniques, such as classifier guidance (CG) and classifier-free guidance (CFG). These techniques are often not applicable in unconditional generation or in various downstream tasks such as image restoration. In this paper, we propose a novel sampling guidance, called Perturbed-Attention Guidance (PAG), which improves diffusion sample quality across both unconditional and conditional settings, achieving this without requiring additional training or the integration of external modules. PAG is designed to progressively enhance the structure of samples throughout the denoising process. It involves generating intermediate samples with degraded structure by substituting selected self-attention maps in diffusion U-Net with an identity matrix, by considering the self-attention mechanisms' ability to capture structural information, and guiding the denoising process away from these degraded samples. In both ADM and Stable Diffusion, PAG surprisingly improves sample quality in conditional and even unconditional scenarios. Moreover, PAG significantly improves the baseline performance in various downstream tasks where existing guidances such as CG or CFG cannot be fully utilized, including ControlNet with empty prompts and image restoration such as inpainting and deblurring.

comment: Project page is available at https://ku-cvlab.github.io/Perturbed-Attention-Guidance

☆ AIDE: An Automatic Data Engine for Object Detection in Autonomous Driving CVPR-2024

Mingfu Liang, Jong-Chyi Su, Samuel Schulter, Sparsh Garg, Shiyu Zhao, Ying Wu, Manmohan Chandraker

Autonomous vehicle (AV) systems rely on robust perception models as a cornerstone of safety assurance. However, objects encountered on the road exhibit a long-tailed distribution, with rare or unseen categories posing challenges to a deployed perception model. This necessitates an expensive process of continuously curating and annotating data with significant human effort. We propose to leverage recent advances in vision-language and large language models to design an Automatic Data Engine (AIDE) that automatically identifies issues, efficiently curates data, improves the model through auto-labeling, and verifies the model through generation of diverse scenarios. This process operates iteratively, allowing for continuous self-improvement of the model. We further establish a benchmark for open-world detection on AV datasets to comprehensively evaluate various learning paradigms, demonstrating our method's superior performance at a reduced cost.

comment: Accepted by CVPR-2024

☆ ChatGPT Rates Natural Language Explanation Quality Like Humans: But on Which Scales? LREC

Fan Huang, Haewoon Kwak, Kunwoo Park, Jisun An

As AI becomes more integral in our lives, the need for transparency and responsibility grows. While natural language explanations (NLEs) are vital for clarifying the reasoning behind AI decisions, evaluating them through human judgments is complex and resource-intensive due to subjectivity and the need for fine-grained ratings. This study explores the alignment between ChatGPT and human assessments across multiple scales (i.e., binary, ternary, and 7-Likert scale). We sample 300 data instances from three NLE datasets and collect 900 human annotations for both informativeness and clarity scores as the text quality measurement. We further conduct paired comparison experiments under different ranges of subjectivity scores, where the baseline comes from 8,346 human annotations. Our results show that ChatGPT aligns better with humans in more coarse-grained scales. Also, paired comparisons and dynamic prompting (i.e., providing semantically similar examples in the prompt) improve the alignment. This research advances our understanding of large language models' capabilities to assess the text explanation quality in different configurations for responsible AI development.

comment: Accpeted by LREC-COLING 2024 main conference, long paper

☆ Bridging Textual and Tabular Worlds for Fact Verification: A Lightweight, Attention-Based Model LREC

Shirin Dabbaghi Varnosfaderani, Canasai Kruengkrai, Ramin Yahyapour, Junichi Yamagishi

FEVEROUS is a benchmark and research initiative focused on fact extraction and verification tasks involving unstructured text and structured tabular data. In FEVEROUS, existing works often rely on extensive preprocessing and utilize rule-based transformations of data, leading to potential context loss or misleading encodings. This paper introduces a simple yet powerful model that nullifies the need for modality conversion, thereby preserving the original evidence's context. By leveraging pre-trained models on diverse text and tabular datasets and by incorporating a lightweight attention-based mechanism, our approach efficiently exploits latent connections between different data types, thereby yielding comprehensive and reliable verdict predictions. The model's modular structure adeptly manages multi-modal information, ensuring the integrity and authenticity of the original evidence are uncompromised. Comparative analyses reveal that our approach exhibits competitive performance, aligning itself closely with top-tier models on the FEVEROUS benchmark.

comment: Accepted for a presentation at LREC-COLING 2024 - The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation

☆ Addressing Myopic Constrained POMDP Planning with Recursive Dual Ascent

Paula Stocco, Suhas Chundi, Arec Jamgochian, Mykel J. Kochenderfer

Lagrangian-guided Monte Carlo tree search with global dual ascent has been applied to solve large constrained partially observable Markov decision processes (CPOMDPs) online. In this work, we demonstrate that these global dual parameters can lead to myopic action selection during exploration, ultimately leading to suboptimal decision making. To address this, we introduce history-dependent dual variables that guide local action selection and are optimized with recursive dual ascent. We empirically compare the performance of our approach on a motivating toy example and two large CPOMDPs, demonstrating improved exploration, and ultimately, safer outcomes.

comment: Accepted to the 2024 International Conference on Automated Planning and Scheduling (ICAPS)

☆ MESIA: Understanding and Leveraging Supplementary Nature of Method-level Comments for Automatic Comment Generation

Xinglu Pan, Chenxiao Liu, Yanzhen Zou, Tao Xie, Bing Xie

Code comments are important for developers in program comprehension. In scenarios of comprehending and reusing a method, developers expect code comments to provide supplementary information beyond the method signature. However, the extent of such supplementary information varies a lot in different code comments. In this paper, we raise the awareness of the supplementary nature of method-level comments and propose a new metric named MESIA (Mean Supplementary Information Amount) to assess the extent of supplementary information that a code comment can provide. With the MESIA metric, we conduct experiments on a popular code-comment dataset and three common types of neural approaches to generate method-level comments. Our experimental results demonstrate the value of our proposed work with a number of findings. (1) Small-MESIA comments occupy around 20% of the dataset and mostly fall into only the WHAT comment category. (2) Being able to provide various kinds of essential information, large-MESIA comments in the dataset are difficult for existing neural approaches to generate. (3) We can improve the capability of existing neural approaches to generate large-MESIA comments by reducing the proportion of small-MESIA comments in the training set. (4) The retrained model can generate large-MESIA comments that convey essential meaningful supplementary information for methods in the small-MESIA test set, but will get a lower BLEU score in evaluation. These findings indicate that with good training data, auto-generated comments can sometimes even surpass human-written reference comments, and having no appropriate ground truth for evaluation is an issue that needs to be addressed by future work on automatic comment generation.

comment: In 32nd IEEE/ACM International Conference on Program Comprehension (ICPC'24)

☆ The Solution of the Zodiac Killer's 340-Character Cipher

David Oranchak, Sam Blake, Jarl Van Eycke

The case of the Zodiac Killer is one of the most widely known unsolved serial killer cases in history. The unidentified killer murdered five known victims and terrorized the state of California. He also communicated extensively with the press and law enforcement. Besides his murders, Zodiac was known for his use of ciphers. The first Zodiac cipher was solved within a week of its publication, while the second cipher was solved by the authors after 51 years, when it was discovered to be a transposition and homophonic substitution cipher with unusual qualities. In this paper, we detail the historical significance of this cipher and the numerous efforts which culminated in its solution.

☆ The Solution for the ICCV 2023 1st Scientific Figure Captioning Challenge

Dian Chao, Xin Song, Shupeng Zhong, Boyuan Wang, Xiangyu Wu, Chen Zhu, Yang Yang

In this paper, we propose a solution for improving the quality of captions generated for figures in papers. We adopt the approach of summarizing the textual content in the paper to generate image captions. Throughout our study, we encounter discrepancies in the OCR information provided in the official dataset. To rectify this, we employ the PaddleOCR toolkit to extract OCR information from all images. Moreover, we observe that certain textual content in the official paper pertains to images that are not relevant for captioning, thereby introducing noise during caption generation. To mitigate this issue, we leverage LLaMA to extract image-specific information by querying the textual content based on image mentions, effectively filtering out extraneous information. Additionally, we recognize a discrepancy between the primary use of maximum likelihood estimation during text generation and the evaluation metrics such as ROUGE employed to assess the quality of generated captions. To bridge this gap, we integrate the BRIO model framework, enabling a more coherent alignment between the generation and evaluation processes. Our approach ranked first in the final test with a score of 4.49.

☆ Reinforcement Learning-based Receding Horizon Control using Adaptive Control Barrier Functions for Safety-Critical Systems

Ehsan Sabouni, H. M. Sabbir Ahmad, Vittorio Giammarino, Christos G. Cassandras, Ioannis Ch. Paschalidis, Wenchao Li

Optimal control methods provide solutions to safety-critical problems but easily become intractable. Control Barrier Functions (CBFs) have emerged as a popular technique that facilitates their solution by provably guaranteeing safety, through their forward invariance property, at the expense of some performance loss. This approach involves defining a performance objective alongside CBF-based safety constraints that must always be enforced. Unfortunately, both performance and solution feasibility can be significantly impacted by two key factors: (i) the selection of the cost function and associated parameters, and (ii) the calibration of parameters within the CBF-based constraints, which capture the trade-off between performance and conservativeness. %as well as infeasibility. To address these challenges, we propose a Reinforcement Learning (RL)-based Receding Horizon Control (RHC) approach leveraging Model Predictive Control (MPC) with CBFs (MPC-CBF). In particular, we parameterize our controller and use bilevel optimization, where RL is used to learn the optimal parameters while MPC computes the optimal control input. We validate our method by applying it to the challenging automated merging control problem for Connected and Automated Vehicles (CAVs) at conflicting roadways. Results demonstrate improved performance and a significant reduction in the number of infeasible cases compared to traditional heuristic approaches used for tuning CBF-based controllers, showcasing the effectiveness of the proposed method.

☆ The Pursuit of Fairness in Artificial Intelligence Models: A Survey

Tahsin Alamgir Kheya, Mohamed Reda Bouadjenek, Sunil Aryal

Artificial Intelligence (AI) models are now being utilized in all facets of our lives such as healthcare, education and employment. Since they are used in numerous sensitive environments and make decisions that can be life altering, potential biased outcomes are a pressing matter. Developers should ensure that such models don't manifest any unexpected discriminatory practices like partiality for certain genders, ethnicities or disabled people. With the ubiquitous dissemination of AI systems, researchers and practitioners are becoming more aware of unfair models and are bound to mitigate bias in them. Significant research has been conducted in addressing such issues to ensure models don't intentionally or unintentionally perpetuate bias. This survey offers a synopsis of the different ways researchers have promoted fairness in AI systems. We explore the different definitions of fairness existing in the current literature. We create a comprehensive taxonomy by categorizing different types of bias and investigate cases of biased AI in different application domains. A thorough study is conducted of the approaches and techniques employed by researchers to mitigate bias in AI models. Moreover, we also delve into the impact of biased models on user experience and the ethical considerations to contemplate when developing and deploying such models. We hope this survey helps researchers and practitioners understand the intricate details of fairness and bias in AI systems. By sharing this thorough survey, we aim to promote additional discourse in the domain of equitable and responsible AI.

comment: 37 pages, 6 figures

☆ Deep Support Vectors

Junhoo Lee, Hyunho Lee, Kyomin Hwang, Nojun Kwak

While the success of deep learning is commonly attributed to its theoretical equivalence with Support Vector Machines (SVM), the practical implications of this relationship have not been thoroughly explored. This paper pioneers an exploration in this domain, specifically focusing on the identification of Deep Support Vectors (DSVs) within deep learning models. We introduce the concept of DeepKKT conditions, an adaptation of the traditional Karush-Kuhn-Tucker (KKT) conditions tailored for deep learning. Through empirical investigations, we illustrate that DSVs exhibit similarities to support vectors in SVM, offering a tangible method to interpret the decision-making criteria of models. Additionally, our findings demonstrate that models can be effectively reconstructed using DSVs, resembling the process in SVM. The code will be available.

☆ Learning Traffic Signal Control via Genetic Programming

Xiao-Cheng Liao, Yi Mei, Mengjie Zhang

The control of traffic signals is crucial for improving transportation efficiency. Recently, learning-based methods, especially Deep Reinforcement Learning (DRL), garnered substantial success in the quest for more efficient traffic signal control strategies. However, the design of rewards in DRL highly demands domain knowledge to converge to an effective policy, and the final policy also presents difficulties in terms of explainability. In this work, a new learning-based method for signal control in complex intersections is proposed. In our approach, we design a concept of phase urgency for each signal phase. During signal transitions, the traffic light control strategy selects the next phase to be activated based on the phase urgency. We then proposed to represent the urgency function as an explainable tree structure. The urgency function can calculate the phase urgency for a specific phase based on the current road conditions. Genetic programming is adopted to perform gradient-free optimization of the urgency function. We test our algorithm on multiple public traffic signal control datasets. The experimental results indicate that the tree-shaped urgency function evolved by genetic programming outperforms the baselines, including a state-of-the-art method in the transportation field and a well-known DRL-based method.

☆ JMultiWOZ: A Large-Scale Japanese Multi-Domain Task-Oriented Dialogue Dataset LREC

Atsumoto Ohashi, Ryu Hirai, Shinya Iizuka, Ryuichiro Higashinaka

Dialogue datasets are crucial for deep learning-based task-oriented dialogue system research. While numerous English language multi-domain task-oriented dialogue datasets have been developed and contributed to significant advancements in task-oriented dialogue systems, such a dataset does not exist in Japanese, and research in this area is limited compared to that in English. In this study, towards the advancement of research and development of task-oriented dialogue systems in Japanese, we constructed JMultiWOZ, the first Japanese language large-scale multi-domain task-oriented dialogue dataset. Using JMultiWOZ, we evaluated the dialogue state tracking and response generation capabilities of the state-of-the-art methods on the existing major English benchmark dataset MultiWOZ2.2 and the latest large language model (LLM)-based methods. Our evaluation results demonstrated that JMultiWOZ provides a benchmark that is on par with MultiWOZ2.2. In addition, through evaluation experiments of interactive dialogues with the models and human participants, we identified limitations in the task completion capabilities of LLMs in Japanese.

comment: Accepted by LREC-COLING 2024

☆ ALISA: Accelerating Large Language Model Inference via Sparsity-Aware KV Caching ISCA 2024

Youpeng Zhao, Di Wu, Jun Wang

The Transformer architecture has significantly advanced natural language processing (NLP) and has been foundational in developing large language models (LLMs) such as LLaMA and OPT, which have come to dominate a broad range of NLP tasks. Despite their superior accuracy, LLMs present unique challenges in practical inference, concerning the compute and memory-intensive nature. Thanks to the autoregressive characteristic of LLM inference, KV caching for the attention layers in Transformers can effectively accelerate LLM inference by substituting quadratic-complexity computation with linear-complexity memory accesses. Yet, this approach requires increasing memory as demand grows for processing longer sequences. The overhead leads to reduced throughput due to I/O bottlenecks and even out-of-memory errors, particularly on resource-constrained systems like a single commodity GPU. In this paper, we propose ALISA, a novel algorithm-system co-design solution to address the challenges imposed by KV caching. On the algorithm level, ALISA prioritizes tokens that are most important in generating a new token via a Sparse Window Attention (SWA) algorithm. SWA introduces high sparsity in attention layers and reduces the memory footprint of KV caching at negligible accuracy loss. On the system level, ALISA employs three-phase token-level dynamical scheduling and optimizes the trade-off between caching and recomputation, thus maximizing the overall performance in resource-constrained systems. In a single GPU-CPU system, we demonstrate that under varying workloads, ALISA improves the throughput of baseline systems such as FlexGen and vLLM by up to 3X and 1.9X, respectively.

comment: ISCA 2024

☆ Neural Multimodal Topic Modeling: A Comprehensive Evaluation LREC

Felipe González-Pizarro, Giuseppe Carenini

Neural topic models can successfully find coherent and diverse topics in textual data. However, they are limited in dealing with multimodal datasets (e.g., images and text). This paper presents the first systematic and comprehensive evaluation of multimodal topic modeling of documents containing both text and images. In the process, we propose two novel topic modeling solutions and two novel evaluation metrics. Overall, our evaluation on an unprecedented rich and diverse collection of datasets indicates that both of our models generate coherent and diverse topics. Nevertheless, the extent to which one method outperforms the other depends on the metrics and dataset combinations, which suggests further exploration of hybrid solutions in the future. Notably, our succinct human evaluation aligns with the outcomes determined by our proposed metrics. This alignment not only reinforces the credibility of our metrics but also highlights the potential for their application in guiding future multimodal topic modeling endeavors.

comment: Camera-Ready for LREC-COLING 2024 (Long Paper)

☆ Visual Hallucination: Definition, Quantification, and Prescriptive Remediations

Vipula Rawte, Anku Rani, Harshad Sharma, Neeraj Anand, Krishnav Rajbangshi, Amit Sheth, Amitava Das

The troubling rise of hallucination presents perhaps the most significant impediment to the advancement of responsible AI. In recent times, considerable research has focused on detecting and mitigating hallucination in Large Language Models (LLMs). However, it's worth noting that hallucination is also quite prevalent in Vision-Language models (VLMs). In this paper, we offer a fine-grained discourse on profiling VLM hallucination based on two tasks: i) image captioning, and ii) Visual Question Answering (VQA). We delineate eight fine-grained orientations of visual hallucination: i) Contextual Guessing, ii) Identity Incongruity, iii) Geographical Erratum, iv) Visual Illusion, v) Gender Anomaly, vi) VLM as Classifier, vii) Wrong Reading, and viii) Numeric Discrepancy. We curate Visual HallucInation eLiciTation (VHILT), a publicly available dataset comprising 2,000 samples generated using eight VLMs across two tasks of captioning and VQA along with human annotations for the categories as mentioned earlier.

☆ InternLM2 Technical Report

Zheng Cai, Maosong Cao, Haojiong Chen, Kai Chen, Keyu Chen, Xin Chen, Xun Chen, Zehui Chen, Zhi Chen, Pei Chu, Xiaoyi Dong, Haodong Duan, Qi Fan, Zhaoye Fei, Yang Gao, Jiaye Ge, Chenya Gu, Yuzhe Gu, Tao Gui, Aijia Guo, Qipeng Guo, Conghui He, Yingfan Hu, Ting Huang, Tao Jiang, Penglong Jiao, Zhenjiang Jin, Zhikai Lei, Jiaxing Li, Jingwen Li, Linyang Li, Shuaibin Li, Wei Li, Yining Li, Hongwei Liu, Jiangning Liu, Jiawei Hong, Kaiwen Liu, Kuikun Liu, Xiaoran Liu, Chengqi Lv, Haijun Lv, Kai Lv, Li Ma, Runyuan Ma, Zerun Ma, Wenchang Ning, Linke Ouyang, Jiantao Qiu, Yuan Qu, Fukai Shang, Yunfan Shao, Demin Song, Zifan Song, Zhihao Sui, Peng Sun, Yu Sun, Huanze Tang, Bin Wang, Guoteng Wang, Jiaqi Wang, Jiayu Wang, Rui Wang, Yudong Wang, Ziyi Wang, Xingjian Wei, Qizhen Weng, Fan Wu, Yingtong Xiong, Chao Xu, Ruiliang Xu, Hang Yan, Yirong Yan, Xiaogui Yang, Haochen Ye, Huaiyuan Ying, Jia Yu, Jing Yu, Yuhang Zang, Chuyu Zhang, Li Zhang, Pan Zhang, Peng Zhang, Ruijie Zhang, Shuo Zhang, Songyang Zhang, Wenjian Zhang, Wenwei Zhang, Xingcheng Zhang, Xinyue Zhang, Hui Zhao, Qian Zhao, Xiaomeng Zhao, Fengzhe Zhou, Zaida Zhou, Jingming Zhuo, Yicheng Zou, Xipeng Qiu, Yu Qiao, Dahua Lin

The evolution of Large Language Models (LLMs) like ChatGPT and GPT-4 has sparked discussions on the advent of Artificial General Intelligence (AGI). However, replicating such advancements in open-source models has been challenging. This paper introduces InternLM2, an open-source LLM that outperforms its predecessors in comprehensive evaluations across 6 dimensions and 30 benchmarks, long-context modeling, and open-ended subjective evaluations through innovative pre-training and optimization techniques. The pre-training process of InternLM2 is meticulously detailed, highlighting the preparation of diverse data types including text, code, and long-context data. InternLM2 efficiently captures long-term dependencies, initially trained on 4k tokens before advancing to 32k tokens in pre-training and fine-tuning stages, exhibiting remarkable performance on the 200k ``Needle-in-a-Haystack" test. InternLM2 is further aligned using Supervised Fine-Tuning (SFT) and a novel Conditional Online Reinforcement Learning from Human Feedback (COOL RLHF) strategy that addresses conflicting human preferences and reward hacking. By releasing InternLM2 models in different training stages and model sizes, we provide the community with insights into the model's evolution.

♻ ☆ CLHA: A Simple yet Effective Contrastive Learning Framework for Human Alignment

Feiteng Fang, Liang Zhu, Min Yang, Xi Feng, Jinchang Hou, Qixuan Zhao, Chengming Li, Xiping Hu, Ruifeng Xu

Reinforcement learning from human feedback (RLHF) is a crucial technique in aligning large language models (LLMs) with human preferences, ensuring these LLMs behave in beneficial and comprehensible ways to users. However, a longstanding challenge in human alignment techniques based on reinforcement learning lies in their inherent complexity and difficulty in training. To address this challenge, we present a simple yet effective Contrastive Learning Framework for Human Alignment (CLHA) to align LLMs with human preferences directly. CLHA employs a novel rescoring strategy to evaluate the noise within the data by considering its inherent quality and dynamically adjusting the training process. Simultaneously, CLHA utilizes pairwise contrastive loss and adaptive supervised fine-tuning loss to adaptively modify the likelihood of generating responses, ensuring enhanced alignment with human preferences. Using advanced methods, CLHA surpasses other algorithms, showcasing superior performance in terms of reward model scores, automatic evaluations, and human assessments on the widely used ``Helpful and Harmless'' dataset.

♻ ☆ Re2LLM: Reflective Reinforcement Large Language Model for Session-based Recommendation

Ziyan Wang, Yingpeng Du, Zhu Sun, Haoyan Chua, Kaidong Feng, Wenya Wang, Jie Zhang

Large Language Models (LLMs) are emerging as promising approaches to enhance session-based recommendation (SBR), where both prompt-based and fine-tuning-based methods have been widely investigated to align LLMs with SBR. However, the former methods struggle with optimal prompts to elicit the correct reasoning of LLMs due to the lack of task-specific feedback, leading to unsatisfactory recommendations. Although the latter methods attempt to fine-tune LLMs with domain-specific knowledge, they face limitations such as high computational costs and reliance on open-source backbones. To address such issues, we propose a \underline{Re}flective \underline{Re}inforcement \underline{L}arge \underline{L}anguage \underline{M}odel (Re2LLM) for SBR, guiding LLMs to focus on specialized knowledge essential for more accurate recommendations effectively and efficiently. In particular, we first design the Reflective Exploration Module to effectively extract knowledge that is readily understandable and digestible by LLMs. To be specific, we direct LLMs to examine recommendation errors through self-reflection and construct a knowledge base (KB) comprising hints capable of rectifying these errors. To efficiently elicit the correct reasoning of LLMs, we further devise the Reinforcement Utilization Module to train a lightweight retrieval agent. It learns to select hints from the constructed KB based on the task-specific feedback, where the hints can serve as guidance to help correct LLMs reasoning for better recommendations. Extensive experiments on multiple real-world datasets demonstrate that our method consistently outperforms state-of-the-art methods.

comment: 11 pages, 4 figures

♻ ☆ Large Language Models in Biomedical and Health Informatics: A Bibliometric Review

Huizi Yu, Lizhou Fan, Lingyao Li, Jiayan Zhou, Zihui Ma, Lu Xian, Wenyue Hua, Sijia He, Mingyu Jin, Yongfeng Zhang, Ashvin Gandhi, Xin Ma

Large Language Models (LLMs) have rapidly become important tools in Biomedical and Health Informatics (BHI), enabling new ways to analyze data, treat patients, and conduct research. This bibliometric review aims to provide a panoramic view of how LLMs have been used in BHI by examining research articles and collaboration networks from 2022 to 2023. It further explores how LLMs can improve Natural Language Processing (NLP) applications in various BHI areas like medical diagnosis, patient engagement, electronic health record management, and personalized medicine. To do this, our bibliometric review identifies key trends, maps out research networks, and highlights major developments in this fast-moving field. Lastly, it discusses the ethical concerns and practical challenges of using LLMs in BHI, such as data privacy and reliable medical recommendations. Looking ahead, we consider how LLMs could further transform biomedical research as well as healthcare delivery and patient outcomes. This bibliometric review serves as a resource for stakeholders in healthcare, including researchers, clinicians, and policymakers, to understand the current state and future potential of LLMs in BHI.

comment: 50 pages, 7 figures, 4 tables

♻ ☆ Guessing human intentions to avoid dangerous situations in caregiving robots IROS2024

Noé Zapata, Gerardo Pérez, Lucas Bonilla, Pedro Núñez, Pilar Bachiller, Pablo Bustos

For robots to interact socially, they must interpret human intentions and anticipate their potential outcomes accurately. This is particularly important for social robots designed for human care, which may face potentially dangerous situations for people, such as unseen obstacles in their way, that should be avoided. This paper explores the Artificial Theory of Mind (ATM) approach to inferring and interpreting human intentions. We propose an algorithm that detects risky situations for humans, selecting a robot action that removes the danger in real time. We use the simulation-based approach to ATM and adopt the 'like-me' policy to assign intentions and actions to people. Using this strategy, the robot can detect and act with a high rate of success under time-constrained situations. The algorithm has been implemented as part of an existing robotics cognitive architecture and tested in simulation scenarios. Three experiments have been conducted to test the implementation's robustness, precision and real-time response, including a simulated scenario, a human-in-the-loop hybrid configuration and a real-world scenario.

comment: 8 pages, 6 figures. Submitted to IROS2024. For associated mpeg file see https://youtu.be/87UEB8P97KY

♻ ☆ Image Captioning in news report scenario

Tianrui Liu, Qi Cai, Changxin Xu, Bo Hong, Jize Xiong, Yuxin Qiao, Tsungwei Yang

Image captioning strives to generate pertinent captions for specified images, situating itself at the crossroads of Computer Vision (CV) and Natural Language Processing (NLP). This endeavor is of paramount importance with far-reaching applications in recommendation systems, news outlets, social media, and beyond. Particularly within the realm of news reporting, captions are expected to encompass detailed information, such as the identities of celebrities captured in the images. However, much of the existing body of work primarily centers around understanding scenes and actions. In this paper, we explore the realm of image captioning specifically tailored for celebrity photographs, illustrating its broad potential for enhancing news industry practices. This exploration aims to augment automated news content generation, thereby facilitating a more nuanced dissemination of information. Our endeavor shows a broader horizon, enriching the narrative in news reporting through a more intuitive image captioning framework.

comment: 10 pages, 4 figures

♻ ☆ Rumor Detection with a novel graph neural network approach

Tianrui Liu, Qi Cai, Changxin Xu, Bo Hong, Fanghao Ni, Yuxin Qiao, Tsungwei Yang

The wide spread of rumors on social media has caused a negative impact on people's daily life, leading to potential panic, fear, and mental health problems for the public. How to debunk rumors as early as possible remains a challenging problem. Existing studies mainly leverage information propagation structure to detect rumors, while very few works focus on correlation among users that they may coordinate to spread rumors in order to gain large popularity. In this paper, we propose a new detection model, that jointly learns both the representations of user correlation and information propagation to detect rumors on social media. Specifically, we leverage graph neural networks to learn the representations of user correlation from a bipartite graph that describes the correlations between users and source tweets, and the representations of information propagation with a tree structure. Then we combine the learned representations from these two modules to classify the rumors. Since malicious users intend to subvert our model after deployment, we further develop a greedy attack scheme to analyze the cost of three adversarial attacks: graph attack, comment attack, and joint attack. Evaluation results on two public datasets illustrate that the proposed MODEL outperforms the state-of-the-art rumor detection models. We also demonstrate our method performs well for early rumor detection. Moreover, the proposed detection method is more robust to adversarial attacks compared to the best existing method. Importantly, we show that it requires a high cost for attackers to subvert user correlation pattern, demonstrating the importance of considering user correlation for rumor detection.

comment: 10 pages, 5 figures

♻ ☆ Enhancing Zero-Shot Chain-of-Thought Reasoning in Large Language Models through Logic COLING 2024

Xufeng Zhao, Mengdi Li, Wenhao Lu, Cornelius Weber, Jae Hee Lee, Kun Chu, Stefan Wermter

Recent advancements in large language models have showcased their remarkable generalizability across various domains. However, their reasoning abilities still have significant room for improvement, especially when confronted with scenarios requiring multi-step reasoning. Although large language models possess extensive knowledge, their reasoning often fails to effectively utilize this knowledge to establish a coherent thinking paradigm. These models sometimes show hallucinations as their reasoning procedures are unconstrained by logical principles. Aiming at improving the zero-shot chain-of-thought reasoning ability of large language models, we propose LoT (Logical Thoughts), a self-improvement prompting framework that leverages principles rooted in symbolic logic, particularly Reductio ad Absurdum, to systematically verify and rectify the reasoning processes step by step. Experimental evaluations conducted on language tasks in diverse domains, including arithmetic, commonsense, symbolic, causal inference, and social problems, demonstrate the efficacy of enhanced reasoning by logic. The implementation code for LoT can be accessed at: https://github.com/xf-zhao/LoT.

comment: Accepted in COLING 2024. Code see https://github.com/xf-zhao/LoT

♻ ☆ X-Portrait: Expressive Portrait Animation with Hierarchical Motion Attention

You Xie, Hongyi Xu, Guoxian Song, Chao Wang, Yichun Shi, Linjie Luo

We propose X-Portrait, an innovative conditional diffusion model tailored for generating expressive and temporally coherent portrait animation. Specifically, given a single portrait as appearance reference, we aim to animate it with motion derived from a driving video, capturing both highly dynamic and subtle facial expressions along with wide-range head movements. As its core, we leverage the generative prior of a pre-trained diffusion model as the rendering backbone, while achieve fine-grained head pose and expression control with novel controlling signals within the framework of ControlNet. In contrast to conventional coarse explicit controls such as facial landmarks, our motion control module is learned to interpret the dynamics directly from the original driving RGB inputs. The motion accuracy is further enhanced with a patch-based local control module that effectively enhance the motion attention to small-scale nuances like eyeball positions. Notably, to mitigate the identity leakage from the driving signals, we train our motion control modules with scaling-augmented cross-identity images, ensuring maximized disentanglement from the appearance reference modules. Experimental results demonstrate the universal effectiveness of X-Portrait across a diverse range of facial portraits and expressive driving sequences, and showcase its proficiency in generating captivating portrait animations with consistently maintained identity characteristics.

♻ ☆ Towards a RAG-based Summarization Agent for the Electron-Ion Collider

Karthik Suresh, Neeltje Kackar, Luke Schleck, Cristiano Fanelli

The complexity and sheer volume of information encompassing documents, papers, data, and other resources from large-scale experiments demand significant time and effort to navigate, making the task of accessing and utilizing these varied forms of information daunting, particularly for new collaborators and early-career scientists. To tackle this issue, a Retrieval Augmented Generation (RAG)--based Summarization AI for EIC (RAGS4EIC) is under development. This AI-Agent not only condenses information but also effectively references relevant responses, offering substantial advantages for collaborators. Our project involves a two-step approach: first, querying a comprehensive vector database containing all pertinent experiment information; second, utilizing a Large Language Model (LLM) to generate concise summaries enriched with citations based on user queries and retrieved data. We describe the evaluation methods that use RAG assessments (RAGAs) scoring mechanisms to assess the effectiveness of responses. Furthermore, we describe the concept of prompt template-based instruction-tuning which provides flexibility and accuracy in summarization. Importantly, the implementation relies on LangChain, which serves as the foundation of our entire workflow. This integration ensures efficiency and scalability, facilitating smooth deployment and accessibility for various user groups within the Electron Ion Collider (EIC) community. This innovative AI-driven framework not only simplifies the understanding of vast datasets but also encourages collaborative participation, thereby empowering researchers. As a demonstration, a web application has been developed to explain each stage of the RAG Agent development in detail.

comment: updated title to have no latex formatting

♻ ☆ Simple and Scalable Strategies to Continually Pre-train Large Language Models

Adam Ibrahim, Benjamin Thérien, Kshitij Gupta, Mats L. Richter, Quentin Anthony, Timothée Lesort, Eugene Belilovsky, Irina Rish

Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution is to continually pre-train these models, saving significant compute compared to re-training. However, the distribution shift induced by new data typically results in degraded performance on previous data or poor adaptation to the new data. In this work, we show that a simple and scalable combination of learning rate (LR) re-warming, LR re-decaying, and replay of previous data is sufficient to match the performance of fully re-training from scratch on all available data, as measured by the final loss and the average score on several language model (LM) evaluation benchmarks. Specifically, we show this for a weak but realistic distribution shift between two commonly used LLM pre-training datasets (English$\rightarrow$English) and a stronger distribution shift (English$\rightarrow$German) at the $405$M parameter model scale with large dataset sizes (hundreds of billions of tokens). Selecting the weak but realistic shift for larger-scale experiments, we also find that our continual learning strategies match the re-training baseline for a 10B parameter LLM. Our results demonstrate that LLMs can be successfully updated via simple and scalable continual learning strategies, matching the re-training baseline using only a fraction of the compute. Finally, inspired by previous work, we propose alternatives to the cosine learning rate schedule that help circumvent forgetting induced by LR re-warming and that are not bound to a fixed token budget.

♻ ☆ DiVa-360: The Dynamic Visual Dataset for Immersive Neural Fields

Cheng-You Lu, Peisen Zhou, Angela Xing, Chandradeep Pokhariya, Arnab Dey, Ishaan Shah, Rugved Mavidipalli, Dylan Hu, Andrew Comport, Kefan Chen, Srinath Sridhar

Advances in neural fields are enabling high-fidelity capture of the shape and appearance of dynamic 3D scenes. However, their capabilities lag behind those offered by conventional representations such as 2D videos because of algorithmic challenges and the lack of large-scale multi-view real-world datasets. We address the dataset limitation with DiVa-360, a real-world 360 dynamic visual dataset that contains synchronized high-resolution and long-duration multi-view video sequences of table-scale scenes captured using a customized low-cost system with 53 cameras. It contains 21 object-centric sequences categorized by different motion types, 25 intricate hand-object interaction sequences, and 8 long-duration sequences for a total of 17.4 M image frames. In addition, we provide foreground-background segmentation masks, synchronized audio, and text descriptions. We benchmark the state-of-the-art dynamic neural field methods on DiVa-360 and provide insights about existing methods and future challenges on long-duration neural field capture.

♻ ☆ Dynamics of Moral Behavior in Heterogeneous Populations of Learning Agents

Elizaveta Tennant, Stephen Hailes, Mirco Musolesi

Growing concerns about safety and alignment of AI systems highlight the importance of embedding moral capabilities in artificial agents. A promising solution is the use of learning from experience, i.e., Reinforcement Learning. In multi-agent (social) environments, complex population-level phenomena may emerge from interactions between individual learning agents. Many of the existing studies rely on simulated social dilemma environments to study the interactions of independent learning agents. However, they tend to ignore the moral heterogeneity that is likely to be present in societies of agents in practice. For example, at different points in time a single learning agent may face opponents who are consequentialist (i.e., caring about maximizing some outcome over time) or norm-based (i.e., focusing on conforming to a specific norm here and now). The extent to which agents' co-development may be impacted by such moral heterogeneity in populations is not well understood. In this paper, we present a study of the learning dynamics of morally heterogeneous populations interacting in a social dilemma setting. Using a Prisoner's Dilemma environment with a partner selection mechanism, we investigate the extent to which the prevalence of diverse moral agents in populations affects individual agents' learning behaviors and emergent population-level outcomes. We observe several types of non-trivial interactions between pro-social and anti-social agents, and find that certain classes of moral agents are able to steer selfish agents towards more cooperative behavior.

♻ ☆ Safe Explicable Planning

Akkamahadevi Hanni, Andrew Boateng, Yu Zhang

Human expectations arise from their understanding of others and the world. In the context of human-AI interaction, this understanding may not align with reality, leading to the AI agent failing to meet expectations and compromising team performance. Explicable planning, introduced as a method to bridge this gap, aims to reconcile human expectations with the agent's optimal behavior, facilitating interpretable decision-making. However, an unresolved critical issue is ensuring safety in explicable planning, as it could result in explicable behaviors that are unsafe. To address this, we propose Safe Explicable Planning (SEP), which extends the prior work to support the specification of a safety bound. The goal of SEP is to find behaviors that align with human expectations while adhering to the specified safety criterion. Our approach generalizes the consideration of multiple objectives stemming from multiple models rather than a single model, yielding a Pareto set of safe explicable policies. We present both an exact method, guaranteeing finding the Pareto set, and a more efficient greedy method that finds one of the policies in the Pareto set. Additionally, we offer approximate solutions based on state aggregation to improve scalability. We provide formal proofs that validate the desired theoretical properties of these methods. Evaluation through simulations and physical robot experiments confirms the effectiveness of our approach for safe explicable planning.

♻ ☆ AI and Generative AI for Research Discovery and Summarization

Mark Glickman, Yi Zhang

AI and generative AI tools, including chatbots like ChatGPT that rely on large language models (LLMs), have burst onto the scene this year, creating incredible opportunities to increase work productivity and improve our lives. Statisticians and data scientists have begun experiencing the benefits from the availability of these tools in numerous ways, such as the generation of programming code from text prompts to analyze data or fit statistical models. One area that these tools can make a substantial impact is in research discovery and summarization. Standalone tools and plugins to chatbots are being developed that allow researchers to more quickly find relevant literature than pre-2023 search tools. Furthermore, generative AI tools have improved to the point where they can summarize and extract the key points from research articles in succinct language. Finally, chatbots based on highly parameterized LLMs can be used to simulate abductive reasoning, which provides researchers the ability to make connections among related technical topics, which can also be used for research discovery. We review the developments in AI and generative AI for research discovery and summarization, and propose directions where these types of tools are likely to head in the future that may be of interest to statistician and data scientists.

comment: 29 pages, 9 figures

♻ ☆ ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review

Sunder Ali Khowaja, Parus Khuwaja, Kapal Dev, Weizheng Wang, Lewis Nkenyereye

ChatGPT is another large language model (LLM) vastly available for the consumers on their devices but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail the issues and concerns raised over chatGPT in line with aforementioned characteristics. We also discuss the recent EU AI Act briefly in accordance with the SPADE evaluation. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for AI policy act, if designed by the governments.

comment: 29 pages, 8 figures, 4 tables

♻ ☆ Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics

Shan Jia, Reilin Lyu, Kangran Zhao, Yize Chen, Zhiyuan Yan, Yan Ju, Chuanbo Hu, Xin Li, Baoyuan Wu, Siwei Lyu

DeepFakes, which refer to AI-generated media content, have become an increasing concern due to their use as a means for disinformation. Detecting DeepFakes is currently solved with programmed machine learning algorithms. In this work, we investigate the capabilities of multimodal large language models (LLMs) in DeepFake detection. We conducted qualitative and quantitative experiments to demonstrate multimodal LLMs and show that they can expose AI-generated images through careful experimental design and prompt engineering. This is interesting, considering that LLMs are not inherently tailored for media forensic tasks, and the process does not require programming. We discuss the limitations of multimodal LLMs for these tasks and suggest possible improvements.

♻ ☆ Efficient Pre-training for Localized Instruction Generation of Videos

Anil Batra, Davide Moltisanti, Laura Sevilla-Lara, Marcus Rohrbach, Frank Keller

Procedural videos show step-by-step demonstrations of tasks like recipe preparation. Understanding such videos is challenging, involving the precise localization of steps and the generation of textual instructions. Manually annotating steps and writing instructions is costly, which limits the size of current datasets and hinders effective learning. Leveraging large but noisy video-transcript datasets for pre-training can boost performance, but demands significant computational resources. Furthermore, transcripts contain irrelevant content and exhibit style variation compared to instructions written by human annotators. To mitigate both issues, we propose a technique, Sieve-&-Swap, to automatically curate a smaller dataset: (i) Sieve filters irrelevant transcripts and (ii) Swap enhances the quality of the text instruction by automatically replacing the transcripts with human-written instructions from a text-only recipe dataset. The curated dataset, three orders of magnitude smaller than current web-scale datasets, enables efficient training of large-scale models with competitive performance. We complement our Sieve-\&-Swap approach with a Procedure Transformer (ProcX) for end-to-end step localization and instruction generation for procedural videos. When this model is pre-trained on our curated dataset, it achieves state-of-the-art performance in zero-shot and finetuning settings on YouCook2 and Tasty, while using a fraction of the computational resources.

comment: This version has some missing experiments and elaborative technical details

♻ ☆ Analyzing the Quality Attributes of AI Vision Models in Open Repositories Under Adversarial Attacks

Zerui Wang, Yan Liu

As AI models rapidly evolve, they are frequently released to open repositories, such as HuggingFace. It is essential to perform quality assurance validation on these models before integrating them into the production development lifecycle. In addition to evaluating efficiency in terms of balanced accuracy and computing costs, adversarial attacks are potential threats to the robustness and explainability of AI models. Meanwhile, XAI applies algorithms that approximate inputs to outputs post-hoc to identify the contributing features. Adversarial perturbations may also degrade the utility of XAI explanations that require further investigation. In this paper, we present an integrated process designed for downstream evaluation tasks, including validating AI model accuracy, evaluating robustness with benchmark perturbations, comparing explanation utility, and assessing overhead. We demonstrate an evaluation scenario involving six computer vision models, which include CNN-based, Transformer-based, and hybrid architectures, three types of perturbations, and five XAI methods, resulting in ninety unique combinations. The process reveals the explanation utility among the XAI methods in terms of the identified key areas responding to the adversarial perturbation. The process produces aggregated results that illustrate multiple attributes of each AI model.

♻ ☆ Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts for Open-Domain QA?

Hexiang Tan, Fei Sun, Wanli Yang, Yuanzhuo Wang, Qi Cao, Xueqi Cheng

While auxiliary information has become a key to enhancing Large Language Models (LLMs), relatively little is known about how LLMs merge these contexts, specifically contexts generated by LLMs and those retrieved from external sources. To investigate this, we formulate a systematic framework to identify whether LLMs' responses, derived from the integration of generated and retrieved contexts, are attributed to either generated or retrieved contexts. To easily trace the origin of the response, we construct datasets with conflicting contexts, i.e., each question is paired with both generated and retrieved contexts, yet only one of them contains the correct answer. Our experiments reveal a significant bias in several LLMs (GPT-4/3.5 and Llama2) to favor generated contexts, even when they provide incorrect information. We further identify two key factors contributing to this bias: i) contexts generated by LLMs typically show greater similarity to the questions, increasing their likelihood of being selected; ii) the segmentation process used in retrieved contexts disrupts their completeness, thereby hindering their full utilization in LLMs. Our analysis enhances the understanding of how LLMs merge diverse contexts, offering valuable insights for advancing current augmentation methods for LLMs.

♻ ☆ Toward a Theory of Causation for Interpreting Neural Code Models

David N. Palacio, Alejandro Velasco, Nathan Cooper, Alvaro Rodriguez, Kevin Moran, Denys Poshyvanyk

Neural Language Models of Code, or Neural Code Models (NCMs), are rapidly progressing from research prototypes to commercial developer tools. As such, understanding the capabilities and limitations of such models is becoming critical. However, the abilities of these models are typically measured using automated metrics that often only reveal a portion of their real-world performance. While, in general, the performance of NCMs appears promising, currently much is unknown about how such models arrive at decisions. To this end, this paper introduces $do_{code}$, a post hoc interpretability method specific to NCMs that is capable of explaining model predictions. $do_{code}$ is based upon causal inference to enable programming language-oriented explanations. While the theoretical underpinnings of $do_{code}$ are extensible to exploring different model properties, we provide a concrete instantiation that aims to mitigate the impact of spurious correlations by grounding explanations of model behavior in properties of programming languages. To demonstrate the practical benefit of $do_{code}$, we illustrate the insights that our framework can provide by performing a case study on two popular deep learning architectures and ten NCMs. The results of this case study illustrate that our studied NCMs are sensitive to changes in code syntax. All our NCMs, except for the BERT-like model, statistically learn to predict tokens related to blocks of code (\eg brackets, parenthesis, semicolon) with less confounding bias as compared to other programming language constructs. These insights demonstrate the potential of $do_{code}$ as a useful method to detect and facilitate the elimination of confounding bias in NCMs.

comment: Accepted to appear in IEEE Transactions on Software Engineering

♻ ☆ Cyber-Security Knowledge Graph Generation by Hierarchical Nonnegative Matrix Factorization

Ryan Barron, Maksim E. Eren, Manish Bhattarai, Selma Wanna, Nicholas Solovyev, Kim Rasmussen, Boian S. Alexandrov, Charles Nicholas, Cynthia Matuszek

Much of human knowledge in cybersecurity is encapsulated within the ever-growing volume of scientific papers. As this textual data continues to expand, the importance of document organization methods becomes increasingly crucial for extracting actionable insights hidden within large text datasets. Knowledge Graphs (KGs) serve as a means to store factual information in a structured manner, providing explicit, interpretable knowledge that includes domain-specific information from the cybersecurity scientific literature. One of the challenges in constructing a KG from scientific literature is the extraction of ontology from unstructured text. In this paper, we address this topic and introduce a method for building a multi-modal KG by extracting structured ontology from scientific papers. We demonstrate this concept in the cybersecurity domain. One modality of the KG represents observable information from the papers, such as the categories in which they were published or the authors. The second modality uncovers latent (hidden) patterns of text extracted through hierarchical and semantic non-negative matrix factorization (NMF), such as named entities, topics or clusters, and keywords. We illustrate this concept by consolidating more than two million scientific papers uploaded to arXiv into the cyber-domain, using hierarchical and semantic NMF, and by building a cyber-domain-specific KG.

comment: Accepted at IEEE ISDFS

♻ ☆ Decode Neural signal as Speech

Yiqian Yang, Yiqun Duan, Qiang Zhang, Renjing Xu, Hui Xiong

Decoding language from brain dynamics is an important open direction in the realm of brain-computer interface (BCI), especially considering the rapid growth of large language models. Compared to invasive-based signals which require electrode implantation surgery, non-invasive neural signals (e.g. EEG, MEG) have attracted increasing attention considering their safety and generality. However, the exploration is not adequate in three aspects: 1) previous methods mainly focus on EEG but none of the previous works address this problem on MEG with better signal quality; 2) prior works have predominantly used ``teacher-forcing" during generative decoding, which is impractical; 3) prior works are mostly ``BART-based" not fully auto-regressive, which performs better in other sequence tasks. In this paper, we explore the brain-to-text translation of MEG signals in a speech-decoding formation. Here we are the first to investigate a cross-attention-based ``whisper" model for generating text directly from MEG signals without teacher forcing. Our model achieves impressive BLEU-1 scores of 60.30 and 52.89 without pretraining \& teacher-forcing on two major datasets (\textit{GWilliams} and \textit{Schoffelen}). This paper conducts a comprehensive review to understand how speech decoding formation performs on the neural decoding tasks, including pretraining initialization, training \& evaluation set splitting, augmentation, and scaling law.

♻ ☆ As Good As A Coin Toss: Human detection of AI-generated images, videos, audio, and audiovisual stimuli

Di Cooke, Abigail Edwards, Sophia Barkoff, Kathryn Kelly

As synthetic media becomes progressively more realistic and barriers to using it continue to lower, the technology has been increasingly utilized for malicious purposes, from financial fraud to nonconsensual pornography. Today, the principal defense against being misled by synthetic media relies on the ability of the human observer to visually and auditorily discern between real and fake. However, it remains unclear just how vulnerable people actually are to deceptive synthetic media in the course of their day to day lives. We conducted a perceptual study with 1276 participants to assess how accurate people were at distinguishing synthetic images, audio only, video only, and audiovisual stimuli from authentic. To reflect the circumstances under which people would likely encounter synthetic media in the wild, testing conditions and stimuli emulated a typical online platform, while all synthetic media used in the survey was sourced from publicly accessible generative AI technology. We find that overall, participants struggled to meaningfully discern between synthetic and authentic content. We also find that detection performance worsens when the stimuli contains synthetic content as compared to authentic content, images featuring human faces as compared to non face objects, a single modality as compared to multimodal stimuli, mixed authenticity as compared to being fully synthetic for audiovisual stimuli, and features foreign languages as compared to languages the observer is fluent in. Finally, we also find that prior knowledge of synthetic media does not meaningfully impact their detection performance. Collectively, these results indicate that people are highly susceptible to being tricked by synthetic media in their daily lives and that human perceptual detection capabilities can no longer be relied upon as an effective counterdefense.

comment: For study pre-registration, see https://osf.io/fnhr3

♻ ☆ RL$^3$: Boosting Meta Reinforcement Learning via RL inside RL$^2$

Abhinav Bhatia, Samer B. Nashed, Shlomo Zilberstein

Meta reinforcement learning (meta-RL) methods such as RL$^2$ have emerged as promising approaches for learning data-efficient RL algorithms tailored to a given task distribution. However, they show poor asymptotic performance and struggle with out-of-distribution tasks because they rely on sequence models, such as recurrent neural networks or transformers, to process experiences rather than summarize them using general-purpose RL components such as value functions. In contrast, traditional RL algorithms are data-inefficient as they do not use domain knowledge, but they do converge to an optimal policy in the limit. We propose RL$^3$, a principled hybrid approach that incorporates action-values, learned per task through traditional RL, in the inputs to meta-RL. We show that RL$^3$ earns greater cumulative reward in the long term, compared to RL$^2$, while maintaining data-efficiency in the short term, and generalizes better to out-of-distribution tasks. Experiments are conducted on both custom and benchmark discrete domains from the meta-RL literature that exhibit a range of short-term, long-term, and complex dependencies.

♻ ☆ Multi-Objective Optimization for Sparse Deep Multi-Task Learning

S. S. Hotegni, M. Berkemeier, S. Peitz

Different conflicting optimization criteria arise naturally in various Deep Learning scenarios. These can address different main tasks (i.e., in the setting of Multi-Task Learning), but also main and secondary tasks such as loss minimization versus sparsity. The usual approach is a simple weighting of the criteria, which formally only works in the convex setting. In this paper, we present a Multi-Objective Optimization algorithm using a modified Weighted Chebyshev scalarization for training Deep Neural Networks (DNNs) with respect to several tasks. By employing this scalarization technique, the algorithm can identify all optimal solutions of the original problem while reducing its complexity to a sequence of single-objective problems. The simplified problems are then solved using an Augmented Lagrangian method, enabling the use of popular optimization techniques such as Adam and Stochastic Gradient Descent, while efficaciously handling constraints. Our work aims to address the (economical and also ecological) sustainability issue of DNN models, with a particular focus on Deep Multi-Task models, which are typically designed with a very large number of weights to perform equally well on multiple tasks. Through experiments conducted on two Machine Learning datasets, we demonstrate the possibility of adaptively sparsifying the model during training without significantly impacting its performance, if we are willing to apply task-specific adaptations to the network weights. Code is available at https://github.com/salomonhotegni/MDMTN

comment: 12 pages, 7 figures

♻ ☆ MedPromptX: Grounded Multimodal Prompting for Chest X-ray Diagnosis

Mai A. Shaaban, Adnan Khan, Mohammad Yaqub

Chest X-ray images are commonly used for predicting acute and chronic cardiopulmonary conditions, but efforts to integrate them with structured clinical data face challenges due to incomplete electronic health records (EHR). This paper introduces \textbf{MedPromptX}, the first model to integrate multimodal large language models (MLLMs), few-shot prompting (FP) and visual grounding (VG) to combine imagery with EHR data for chest X-ray diagnosis. A pre-trained MLLM is utilized to complement the missing EHR information, providing a comprehensive understanding of patients' medical history. Additionally, FP reduces the necessity for extensive training of MLLMs while effectively tackling the issue of hallucination. Nevertheless, the process of determining the optimal number of few-shot examples and selecting high-quality candidates can be burdensome, yet it profoundly influences model performance. Hence, we propose a new technique that dynamically refines few-shot data for real-time adjustment to new patient scenarios. Moreover, VG aids in focusing the model's attention on relevant regions of interest in X-ray images, enhancing the identification of abnormalities. We release MedPromptX-VQA, a new in-context visual question answering dataset encompassing interleaved image and EHR data derived from MIMIC-IV and MIMIC-CXR databases. Results demonstrate the SOTA performance of MedPromptX, achieving an 11% improvement in F1-score compared to the baselines. Code and data are available at https://github.com/BioMedIA-MBZUAI/MedPromptX

♻ ☆ Accelerating Scientific Discovery with Generative Knowledge Extraction, Graph-Based Representation, and Multimodal Intelligent Graph Reasoning

Markus J. Buehler

Leveraging generative Artificial Intelligence (AI), we have transformed a dataset comprising 1,000 scientific papers into an ontological knowledge graph. Through an in-depth structural analysis, we have calculated node degrees, identified communities and connectivities, and evaluated clustering coefficients and betweenness centrality of pivotal nodes, uncovering fascinating knowledge architectures. The graph has an inherently scale-free nature, is highly connected, and can be used for graph reasoning by taking advantage of transitive and isomorphic properties that reveal unprecedented interdisciplinary relationships that can be used to answer queries, identify gaps in knowledge, propose never-before-seen material designs, and predict material behaviors. We compute deep node embeddings for combinatorial node similarity ranking for use in a path sampling strategy links dissimilar concepts that have previously not been related. One comparison revealed structural parallels between biological materials and Beethoven's 9th Symphony, highlighting shared patterns of complexity through isomorphic mapping. In another example, the algorithm proposed a hierarchical mycelium-based composite based on integrating path sampling with principles extracted from Kandinsky's 'Composition VII' painting. The resulting material integrates an innovative set of concepts that include a balance of chaos/order, adjustable porosity, mechanical strength, and complex patterned chemical functionalization. We uncover other isomorphisms across science, technology and art, revealing a nuanced ontology of immanence that reveal a context-dependent heterarchical interplay of constituents. Graph-based generative AI achieves a far higher degree of novelty, explorative capacity, and technical detail, than conventional approaches and establishes a widely useful framework for innovation by revealing hidden connections.

♻ ☆ Text-Guided Variational Image Generation for Industrial Anomaly Detection and Segmentation CVPR 2024

Mingyu Lee, Jongwon Choi

We propose a text-guided variational image generation method to address the challenge of getting clean data for anomaly detection in industrial manufacturing. Our method utilizes text information about the target object, learned from extensive text library documents, to generate non-defective data images resembling the input image. The proposed framework ensures that the generated non-defective images align with anticipated distributions derived from textual and image-based knowledge, ensuring stability and generality. Experimental results demonstrate the effectiveness of our approach, surpassing previous methods even with limited non-defective data. Our approach is validated through generalization tests across four baseline models and three distinct datasets. We present an additional analysis to enhance the effectiveness of anomaly detection models by utilizing the generated images.

comment: 18 pages, Accepted to CVPR 2024

♻ ☆ Unveiling the Pitfalls of Knowledge Editing for Large Language Models ICLR 2024

Zhoubo Li, Ningyu Zhang, Yunzhi Yao, Mengru Wang, Xi Chen, Huajun Chen

As the cost associated with fine-tuning Large Language Models (LLMs) continues to rise, recent research efforts have pivoted towards developing methodologies to edit implicit knowledge embedded within LLMs. Yet, there's still a dark cloud lingering overhead -- will knowledge editing trigger butterfly effect? since it is still unclear whether knowledge editing might introduce side effects that pose potential risks or not. This paper pioneers the investigation into the potential pitfalls associated with knowledge editing for LLMs. To achieve this, we introduce new benchmark datasets and propose innovative evaluation metrics. Our results underline two pivotal concerns: (1) Knowledge Conflict: Editing groups of facts that logically clash can magnify the inherent inconsistencies in LLMs-a facet neglected by previous methods. (2) Knowledge Distortion: Altering parameters with the aim of editing factual knowledge can irrevocably warp the innate knowledge structure of LLMs. Experimental results vividly demonstrate that knowledge editing might inadvertently cast a shadow of unintended consequences on LLMs, which warrant attention and efforts for future works. Code and data are available at https://github.com/zjunlp/PitfallsKnowledgeEditing.

comment: ICLR 2024

♻ ☆ Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration NAACL 2024

Zhenhailong Wang, Shaoguang Mao, Wenshan Wu, Tao Ge, Furu Wei, Heng Ji

Human intelligence thrives on cognitive synergy, where collaboration among different minds yield superior outcomes compared to isolated individuals. In this work, we propose Solo Performance Prompting (SPP), which transforms a single LLM into a cognitive synergist by engaging in multi-turn self-collaboration with multiple personas. A cognitive synergist is an intelligent agent that collaboratively combines multiple minds' strengths and knowledge to enhance problem-solving in complex tasks. By dynamically identifying and simulating different personas based on task inputs, SPP unleashes the potential of cognitive synergy in LLMs. Our in-depth analysis shows that assigning multiple fine-grained personas in LLMs improves problem-solving abilities compared to using a single or fixed number of personas. We evaluate SPP on three challenging tasks: Trivia Creative Writing, Codenames Collaborative, and Logic Grid Puzzle, encompassing both knowledge-intensive and reasoning-intensive types. Unlike previous works, such as Chain-of-Thought, that solely enhance the reasoning abilities in LLMs, experimental results demonstrate that SPP effectively reduces factual hallucination, and maintains strong reasoning capabilities. Additionally, comparative experiments show that cognitive synergy only emerges in GPT-4 and does not appear in less capable models, such as GPT-3.5-turbo and Llama2-13b-chat, which draws an interesting analogy to human development. Code, data, and prompts can be found at: https://github.com/MikeWangWZHL/Solo-Performance-Prompting.git.

comment: Accepted as a main conference paper at NAACL 2024

♻ ☆ A Safe Preference Learning Approach for Personalization with Applications to Autonomous Vehicles

Ruya Karagulle, Nikos Arechiga, Andrew Best, Jonathan DeCastro, Necmiye Ozay

This work introduces a preference learning method that ensures adherence to given specifications, with an application to autonomous vehicles. Our approach incorporates the priority ordering of Signal Temporal Logic (STL) formulas describing traffic rules into a learning framework. By leveraging Parametric Weighted Signal Temporal Logic (PWSTL), we formulate the problem of safety-guaranteed preference learning based on pairwise comparisons and propose an approach to solve this learning problem. Our approach finds a feasible valuation for the weights of the given PWSTL formula such that, with these weights, preferred signals have weighted quantitative satisfaction measures greater than their non-preferred counterparts. The feasible valuation of weights given by our approach leads to a weighted STL formula that can be used in correct-and-custom-by-construction controller synthesis. We demonstrate the performance of our method with a pilot human subject study in two different simulated driving scenarios involving a stop sign and a pedestrian crossing. Our approach yields competitive results compared to existing preference learning methods in terms of capturing preferences and notably outperforms them when safety is considered.

comment: 9 pages, 3 figures, 2 tables. This work has been published at IEEE Robotics and Automation Letters. Copyright may be transferred without notice, after which this version may no longer be accessible

♻ ☆ Artificial Neural Nets and the Representation of Human Concepts

Timo Freiesleben

What do artificial neural networks (ANNs) learn? The machine learning (ML) community shares the narrative that ANNs must develop abstract human concepts to perform complex tasks. Some go even further and believe that these concepts are stored in individual units of the network. Based on current research, I systematically investigate the assumptions underlying this narrative. I conclude that ANNs are indeed capable of performing complex prediction tasks, and that they may learn human and non-human concepts to do so. However, evidence indicates that ANNs do not represent these concepts in individual units.

comment: For: Philosophy of Science for Machine Learning: Core Issues and New Perspectives, edited by Juan Duran and Giorgia Pozzi

♻ ★ ViT-Lens: Towards Omni-modal Representations CVPR2024

Weixian Lei, Yixiao Ge, Kun Yi, Jianfeng Zhang, Difei Gao, Dylan Sun, Yuying Ge, Ying Shan, Mike Zheng Shou

Aiming to advance AI agents, large foundation models significantly improve reasoning and instruction execution, yet the current focus on vision and language neglects the potential of perceiving diverse modalities in open-world environments. However, the success of data-driven vision and language models is costly or even infeasible to be reproduced for rare modalities. In this paper, we present ViT-Lens-2 that facilitates efficient omni-modal representation learning by perceiving novel modalities with a pretrained ViT and aligning them to a pre-defined space. Specifically, the modality-specific lens is tuned to project any-modal signals to an intermediate embedding space, which are then processed by a strong ViT with pre-trained visual knowledge. The encoded representations are optimized toward aligning with the modal-independent space, pre-defined by off-the-shelf foundation models. ViT-Lens-2 provides a unified solution for representation learning of increasing modalities with two appealing advantages: (i) Unlocking the great potential of pretrained ViTs to novel modalities effectively with efficient data regime; (ii) Enabling emergent downstream capabilities through modality alignment and shared ViT parameters. We tailor ViT-Lens-2 to learn representations for 3D point cloud, depth, audio, tactile and EEG, and set new state-of-the-art results across various understanding tasks, such as zero-shot classification. By seamlessly integrating ViT-Lens-2 into Multimodal Foundation Models, we enable Any-modality to Text and Image Generation in a zero-shot manner. Code and models are available at https://github.com/TencentARC/ViT-Lens.

comment: This work is a follow-up of arXiv:2308.10185. Accepted to CVPR2024

♻ ☆ AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation CVPR 2024

Jeongsoo Choi, Se Jin Park, Minsu Kim, Yong Man Ro

This paper proposes a novel direct Audio-Visual Speech to Audio-Visual Speech Translation (AV2AV) framework, where the input and output of the system are multimodal (i.e., audio and visual speech). With the proposed AV2AV, two key advantages can be brought: 1) We can perform real-like conversations with individuals worldwide in a virtual meeting by utilizing our own primary languages. In contrast to Speech-to-Speech Translation (A2A), which solely translates between audio modalities, the proposed AV2AV directly translates between audio-visual speech. This capability enhances the dialogue experience by presenting synchronized lip movements along with the translated speech. 2) We can improve the robustness of the spoken language translation system. By employing the complementary information of audio-visual speech, the system can effectively translate spoken language even in the presence of acoustic noise, showcasing robust performance. To mitigate the problem of the absence of a parallel AV2AV translation dataset, we propose to train our spoken language translation system with the audio-only dataset of A2A. This is done by learning unified audio-visual speech representations through self-supervised learning in advance to train the translation system. Moreover, we propose an AV-Renderer that can generate raw audio and video in parallel. It is designed with zero-shot speaker modeling, thus the speaker in source audio-visual speech can be maintained at the target translated audio-visual speech. The effectiveness of AV2AV is evaluated with extensive experiments in a many-to-many language translation setting. Demo page is available on https://choijeongsoo.github.io/av2av.

comment: CVPR 2024. Code & Demo: https://choijeongsoo.github.io/av2av

♻ ☆ Towards more Practical Threat Models in Artificial Intelligence Security

Kathrin Grosse, Lukas Bieringer, Tarek Richard Besold, Alexandre Alahi

Recent works have identified a gap between research and practice in artificial intelligence security: threats studied in academia do not always reflect the practical use and security risks of AI. For example, while models are often studied in isolation, they form part of larger ML pipelines in practice. Recent works also brought forward that adversarial manipulations introduced by academic attacks are impractical. We take a first step towards describing the full extent of this disparity. To this end, we revisit the threat models of the six most studied attacks in AI security research and match them to AI usage in practice via a survey with 271 industrial practitioners. On the one hand, we find that all existing threat models are indeed applicable. On the other hand, there are significant mismatches: research is often too generous with the attacker, assuming access to information not frequently available in real-world settings. Our paper is thus a call for action to study more practical threat models in artificial intelligence security.

comment: 18 pages, 4 figures, 8 tables, accepted to Usenix Security, incorporated external feedback

♻ ☆ TagAlign: Improving Vision-Language Alignment with Multi-Tag Classification

Qinying Liu, Wei Wu, Kecheng Zheng, Zhan Tong, Jiawei Liu, Yu Liu, Wei Chen, Zilei Wang, Yujun Shen

The crux of learning vision-language models is to extract semantically aligned information from visual and linguistic data. Existing attempts usually face the problem of coarse alignment, e.g., the vision encoder struggles in localizing an attribute-specified object. In this work, we propose an embarrassingly simple approach to better align image and text features with no need of additional data formats other than image-text pairs. Concretely, given an image and its paired text, we manage to parse objects (e.g., cat) and attributes (e.g., black) from the description, which are highly likely to exist in the image. It is noteworthy that the parsing pipeline is fully automatic and thus enjoys good scalability. With these parsed semantics as supervision signals, we can complement the commonly used image-text contrastive loss with the multi-tag classification loss. Extensive experimental results on a broad suite of semantic segmentation datasets substantiate the average 5.2\% improvement of our framework over existing alternatives. Furthermore, the visualization results indicate that attribute supervision makes vision-language models accurately localize attribute-specified objects. Project page can be found at https://qinying-liu.github.io/Tag-Align.

♻ ☆ SGS-SLAM: Semantic Gaussian Splatting For Neural Dense SLAM

Mingrui Li, Shuhong Liu, Heng Zhou, Guohao Zhu, Na Cheng, Tianchen Deng, Hongyu Wang

We present SGS-SLAM, the first semantic visual SLAM system based on Gaussian Splatting. It incorporates appearance, geometry, and semantic features through multi-channel optimization, addressing the oversmoothing limitations of neural implicit SLAM systems in high-quality rendering, scene understanding, and object-level geometry. We introduce a unique semantic feature loss that effectively compensates for the shortcomings of traditional depth and color losses in object optimization. Through a semantic-guided keyframe selection strategy, we prevent erroneous reconstructions caused by cumulative errors. Extensive experiments demonstrate that SGS-SLAM delivers state-of-the-art performance in camera pose estimation, map reconstruction, precise semantic segmentation, and object-level geometric accuracy, while ensuring real-time rendering capabilities.

♻ ☆ Byzantine-resilient Federated Learning With Adaptivity to Data Heterogeneity

Shiyuan Zuo, Xingrun Yan, Rongfei Fan, Han Hu, Hangguan Shan, Tony Q. S. Quek

This paper deals with federated learning (FL) in the presence of malicious Byzantine attacks and data heterogeneity. A novel Robust Average Gradient Algorithm (RAGA) is proposed, which leverages the geometric median for aggregation and can freely select the round number for local updating. Different from most existing resilient approaches, which perform convergence analysis based on strongly-convex loss function or homogeneously distributed dataset, we conduct convergence analysis for not only strongly-convex but also non-convex loss function over heterogeneous dataset. According to our theoretical analysis, as long as the fraction of dataset from malicious users is less than half, RAGA can achieve convergence at rate $\mathcal{O}({1}/{T^{2/3- \delta}})$ where $T$ is the iteration number and $\delta \in (0, 2/3)$ for non-convex loss function, and at linear rate for strongly-convex loss function. Moreover, stationary point or global optimal solution is proved to obtainable as data heterogeneity vanishes. Experimental results corroborate the robustness of RAGA to Byzantine attacks and verifies the advantage of RAGA over baselines on convergence performance under various intensity of Byzantine attacks, for heterogeneous dataset.

♻ ☆ Hyacinth6B: A large language model for Traditional Chinese

Chih-Wei Song, Yin-Te Tsai

This research's primary motivation of this study is to address the high hardware and computational demands typically associated with LLMs.Therefore,our goal is to find a balance between model lightness and performance,striving to maximize performance while using a comparatively lightweight model. Hyacinth6B was developed with this objective in mind,aiming to fully leverage the core capabilities of LLMs without incurring substantial resource costs, effectively pushing the boundaries of smaller model's performance. The training approach involves parameter efficient finetuning using the LoRA method.

comment: 14pages

♻ ★ Large Language Model for Multi-objective Evolutionary Optimization

Fei Liu, Xi Lin, Zhenkun Wang, Shunyu Yao, Xialiang Tong, Mingxuan Yuan, Qingfu Zhang

Multiobjective evolutionary algorithms (MOEAs) are major methods for solving multiobjective optimization problems (MOPs). Many MOEAs have been proposed in the past decades, of which the search operators need a carefully handcrafted design with domain knowledge. Recently, some attempts have been made to replace the manually designed operators in MOEAs with learning-based operators (e.g., neural network models). However, much effort is still required for designing and training such models, and the learned operators might not generalize well on new problems. To tackle the above challenges, this work investigates a novel approach that leverages the powerful large language model (LLM) to design MOEA operators. With proper prompt engineering, we successfully let a general LLM serve as a black-box search operator for decomposition-based MOEA (MOEA/D) in a zero-shot manner. In addition, by learning from the LLM behavior, we further design an explicit white-box operator with randomness and propose a new version of decomposition-based MOEA, termed MOEA/D-LO. Experimental studies on different test benchmarks show that our proposed method can achieve competitive performance with widely used MOEAs. It is also promising to see the operator only learned from a few instances can have robust generalization performance on unseen problems with quite different patterns and settings. The results reveal the potential benefits of using pre-trained LLMs in the design of MOEAs.To foster reproducibility and accessibility, the source code is https://github.com/FeiLiu36/LLM4MOEA.

♻ ☆ DeepMachining: Online Prediction of Machining Errors of Lathe Machines

Xiang-Li Lu, Hwai-Jung Hsu, Che-Wei Chou, H. T. Kung, Chen-Hsin Lee

We describe DeepMachining, a deep learning-based AI system for online prediction of machining errors of lathe machine operations. We have built and evaluated DeepMachining based on manufacturing data from factories. Specifically, we first pretrain a deep learning model for a given lathe machine's operations to learn the salient features of machining states. Then, we fine-tune the pretrained model to adapt to specific machining tasks. We demonstrate that DeepMachining achieves high prediction accuracy for multiple tasks that involve different workpieces and cutting tools. To the best of our knowledge, this work is one of the first factory experiments using pre-trained deep-learning models to predict machining errors of lathe machines.

♻ ☆ Advancing Topic Segmentation and Outline Generation in Chinese Texts: The Paragraph-level Topic Representation, Corpus, and Benchmark LREC

Feng Jiang, Weihao Liu, Xiaomin Chu, Peifeng Li, Qiaoming Zhu, Haizhou Li

Topic segmentation and outline generation strive to divide a document into coherent topic sections and generate corresponding subheadings, unveiling the discourse topic structure of a document. Compared with sentence-level topic structure, the paragraph-level topic structure can quickly grasp and understand the overall context of the document from a higher level, benefitting many downstream tasks such as summarization, discourse parsing, and information retrieval. However, the lack of large-scale, high-quality Chinese paragraph-level topic structure corpora restrained relative research and applications. To fill this gap, we build the Chinese paragraph-level topic representation, corpus, and benchmark in this paper. Firstly, we propose a hierarchical paragraph-level topic structure representation with three layers to guide the corpus construction. Then, we employ a two-stage man-machine collaborative annotation method to construct the largest Chinese Paragraph-level Topic Structure corpus (CPTS), achieving high quality. We also build several strong baselines, including ChatGPT, to validate the computability of CPTS on two fundamental tasks (topic segmentation and outline generation) and preliminarily verified its usefulness for the downstream task (discourse parsing).

comment: Accepted by LREC-COLING 2024

♻ ☆ ObjectCompose: Evaluating Resilience of Vision-Based Models on Object-to-Background Compositional Changes

Hashmat Shadab Malik, Muhammad Huzaifa, Muzammal Naseer, Salman Khan, Fahad Shahbaz Khan

Given the large-scale multi-modal training of recent vision-based models and their generalization capabilities, understanding the extent of their robustness is critical for their real-world deployment. In this work, we evaluate the resilience of current vision-based models against diverse object-to-background context variations. The majority of robustness evaluation methods have introduced synthetic datasets to induce changes to object characteristics (viewpoints, scale, color) or utilized image transformation techniques (adversarial changes, common corruptions) on real images to simulate shifts in distributions. Recent works have explored leveraging large language models and diffusion models to generate changes in the background. However, these methods either lack in offering control over the changes to be made or distort the object semantics, making them unsuitable for the task. Our method, on the other hand, can induce diverse object-to-background changes while preserving the original semantics and appearance of the object. To achieve this goal, we harness the generative capabilities of text-to-image, image-to-text, and image-to-segment models to automatically generate a broad spectrum of object-to-background changes. We induce both natural and adversarial background changes by either modifying the textual prompts or optimizing the latents and textual embedding of text-to-image models. We produce various versions of standard vision datasets (ImageNet, COCO), incorporating either diverse and realistic backgrounds into the images or introducing color, texture, and adversarial changes in the background. We conduct extensive experiment to analyze the robustness of vision-based models against object-to-background context variations across diverse tasks. Code https://github.com/Muhammad-Huzaifaa/ObjectCompose.git

♻ ☆ Motion Generation from Fine-grained Textual Descriptions

Kunhang Li, Yansong Feng

The task of text2motion is to generate human motion sequences from given textual descriptions, where the model explores diverse mappings from natural language instructions to human body movements. While most existing works are confined to coarse-grained motion descriptions, e.g., "A man squats.", fine-grained descriptions specifying movements of relevant body parts are barely explored. Models trained with coarse-grained texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure to generate motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset specializing in fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with step-by-step instructions with pseudo-code compulsory checks. Accordingly, we design a new text2motion model, FineMotionDiffuse, making full use of fine-grained textual information. Our quantitative evaluation shows that FineMotionDiffuse trained on FineHumanML3D improves FID by a large margin of 0.38, compared with competitive baselines. According to the qualitative evaluation and case study, our model outperforms MotionDiffuse in generating spatially or chronologically composite motions, by learning the implicit mappings from fine-grained descriptions to the corresponding basic motions. We release our data at https://github.com/KunhangL/finemotiondiffuse.

♻ ☆ Tandem Transformers for Inference Efficient LLMs

Aishwarya P S, Pranav Ajit Nair, Yashas Samaga, Toby Boyd, Sanjiv Kumar, Prateek Jain, Praneeth Netrapalli

The autoregressive nature of conventional large language models (LLMs) inherently limits inference speed, as tokens are generated sequentially. While speculative and parallel decoding techniques attempt to mitigate this, they face limitations: either relying on less accurate smaller models for generation or failing to fully leverage the base LLM's representations. We introduce a novel architecture, Tandem transformers, to address these issues. This architecture uniquely combines (1) a small autoregressive model and (2) a large model operating in block mode (processing multiple tokens simultaneously). The small model's predictive accuracy is substantially enhanced by granting it attention to the large model's richer representations. On the PaLM2 pretraining dataset, a tandem of PaLM2-Bison and PaLM2-Gecko demonstrates a 3.3% improvement in next-token prediction accuracy over a standalone PaLM2-Gecko, offering a 1.16x speedup compared to a PaLM2-Otter model with comparable downstream performance. We further incorporate the tandem model within the speculative decoding (SPEED) framework where the large model validates tokens from the small model. This ensures that the Tandem of PaLM2-Bison and PaLM2-Gecko achieves substantial speedup (around 1.14x faster than using vanilla PaLM2-Gecko in SPEED) while maintaining identical downstream task accuracy.

♻ ☆ FedCSD: A Federated Learning Based Approach for Code-Smell Detection

Sadi Alawadi, Khalid Alkharabsheh, Fahed Alkhabbas, Victor Kebande, Feras M. Awaysheh, Fabio Palomba, Mohammed Awad

This paper proposes a Federated Learning Code Smell Detection (FedCSD) approach that allows organizations to collaboratively train federated ML models while preserving their data privacy. These assertions have been supported by three experiments that have significantly leveraged three manually validated datasets aimed at detecting and examining different code smell scenarios. In experiment 1, which was concerned with a centralized training experiment, dataset two achieved the lowest accuracy (92.30%) with fewer smells, while datasets one and three achieved the highest accuracy with a slight difference (98.90% and 99.5%, respectively). This was followed by experiment 2, which was concerned with cross-evaluation, where each ML model was trained using one dataset, which was then evaluated over the other two datasets. Results from this experiment show a significant drop in the model's accuracy (lowest accuracy: 63.80\%) where fewer smells exist in the training dataset, which has a noticeable reflection (technical debt) on the model's performance. Finally, the last and third experiments evaluate our approach by splitting the dataset into 10 companies. The ML model was trained on the company's site, then all model-updated weights were transferred to the server. Ultimately, an accuracy of 98.34% was achieved by the global model that has been trained using 10 companies for 100 training rounds. The results reveal a slight difference in the global model's accuracy compared to the highest accuracy of the centralized model, which can be ignored in favour of the global model's comprehensive knowledge, lower training cost, preservation of data privacy, and avoidance of the technical debt problem.

comment: 17 pages, 7 figures, Journal paper

♻ ☆ Deciphering the Impact of Pretraining Data on Large Language Models through Machine Unlearning

Yang Zhao, Li Du, Xiao Ding, Kai Xiong, Zhouhao Sun, Jun Shi, Ting Liu, Bing Qin

Through pretraining on a corpus with various sources, Large Language Models (LLMs) have gained impressive performance. However, the impact of each component of the pretraining corpus remains opaque. As a result, the organization of the pretraining corpus is still empirical and may deviate from the optimal. To address this issue, we systematically analyze the impact of 48 datasets from 5 major categories of pretraining data of LLMs and measure their impacts on LLMs using benchmarks about nine major categories of model capabilities. Our analyses provide empirical results about the contribution of multiple corpora on the performances of LLMs, along with their joint impact patterns, including complementary, orthogonal, and correlational relationships. We also identify a set of ``high-impact data'' such as Books that is significantly related to a set of model capabilities. These findings provide insights into the organization of data to support more efficient pretraining of LLMs.

♻ ☆ High-throughput Biomedical Relation Extraction for Semi-Structured Web Articles Empowered by Large Language Models

Songchi Zhou, Sheng Yu

Objective: To develop a high-throughput biomedical relation extraction system that takes advantage of the large language models'(LLMs) reading comprehension ability and biomedical world knowledge in a scalable and evidential manner. Methods: We formulate the relation extraction task as binary classifications for large language models. Specifically, LLMs make the decision based on the external corpus and its world knowledge, giving the reason for the judgment for factual verification. This method is tailored for semi-structured web articles, wherein we designate the main title as the tail entity and explicitly incorporate it into the context, and the potential head entities are matched based on a biomedical thesaurus. Moreover, lengthy contents are sliced into text chunks, embedded, and retrieved with additional embedding models. Results: Using an open-source LLM, we extracted 248659 relation triplets of three distinct relation types from three reputable biomedical websites. To assess the efficacy of the basic pipeline employed for biomedical relation extraction, we curated a benchmark dataset annotated by a medical expert. Evaluation results indicate that the pipeline exhibits performance comparable to that of GPT-4. Case studies further illuminate challenges faced by contemporary LLMs in the context of biomedical relation extraction for semi-structured web articles. Conclusion: The proposed method has demonstrated its effectiveness in leveraging the strengths of LLMs for high-throughput biomedical relation extraction. Its adaptability is evident, as it can be seamlessly extended to diverse semi-structured biomedical websites, facilitating the extraction of various types of biomedical relations with ease.

♻ ☆ DreamComposer: Controllable 3D Object Generation via Multi-View Conditions

Yunhan Yang, Yukun Huang, Xiaoyang Wu, Yuan-Chen Guo, Song-Hai Zhang, Hengshuang Zhao, Tong He, Xihui Liu

Utilizing pre-trained 2D large-scale generative models, recent works are capable of generating high-quality novel views from a single in-the-wild image. However, due to the lack of information from multiple views, these works encounter difficulties in generating controllable novel views. In this paper, we present DreamComposer, a flexible and scalable framework that can enhance existing view-aware diffusion models by injecting multi-view conditions. Specifically, DreamComposer first uses a view-aware 3D lifting module to obtain 3D representations of an object from multiple views. Then, it renders the latent features of the target view from 3D representations with the multi-view feature fusion module. Finally the target view features extracted from multi-view inputs are injected into a pre-trained diffusion model. Experiments show that DreamComposer is compatible with state-of-the-art diffusion models for zero-shot novel view synthesis, further enhancing them to generate high-fidelity novel view images with multi-view conditions, ready for controllable 3D object reconstruction and various other applications.

comment: Project Page: https://yhyang-myron.github.io/DreamComposer/

♻ ☆ Attention-based Estimation and Prediction of Human Intent to augment Haptic Glove aided Control of Robotic Hand

Muneeb Ahmed, Rajesh Kumar, Qaim Abbas, Brejesh Lall, Arzad A. Kherani, Sudipto Mukherjee

The letter focuses on Haptic Glove (HG) based control of a Robotic Hand (RH) executing in-hand manipulation of certain objects of interest. The high dimensional motion signals in HG and RH possess intrinsic variability of kinematics resulting in difficulty to establish a direct mapping of the motion signals from HG onto the RH. An estimation mechanism is proposed to quantify the motion signal acquired from the human controller in relation to the intended goal pose of the object being held by the robotic hand. A control algorithm is presented to transform the synthesized intent at the RH and allow relocation of the object to the expected goal pose. The lag in synthesis of the intent in the presence of communication delay leads to a requirement of predicting the estimated intent. We leverage an attention-based convolutional neural network encoder to predict the trajectory of intent for a certain lookahead to compensate for the delays. The proposed methodology is evaluated across objects of different shapes, mass, and materials. We present a comparative performance of the estimation and prediction mechanisms on 5G-driven real-world robotic setup against benchmark methodologies. The test-MSE in prediction of human intent is reported to yield ~ 97.3 -98.7% improvement of accuracy in comparison to LSTM-based benchmark

♻ ☆ Mastering Text, Code and Math Simultaneously via Fusing Highly Specialized Language Models

Ning Ding, Yulin Chen, Ganqu Cui, Xingtai Lv, Weilin Zhao, Ruobing Xie, Bowen Zhou, Zhiyuan Liu, Maosong Sun

Underlying data distributions of natural language, programming code, and mathematical symbols vary vastly, presenting a complex challenge for large language models (LLMs) that strive to achieve high performance across all three domains simultaneously. Achieving a very high level of proficiency for an LLM within a specific domain often requires extensive training with relevant corpora, which is typically accompanied by a sacrifice in performance in other domains. In this paper, we propose to fuse models that are already highly-specialized directly. The proposed fusing framework, UltraFuser, consists of three distinct specialists that are already sufficiently trained on language, coding, and mathematics. A token-level gating mechanism is introduced to blend the specialists' outputs. A two-stage training strategy accompanied by balanced sampling is designed to ensure stability. To effectively train the fused model, we further construct a high-quality supervised instruction tuning dataset, UltraChat 2, which includes text, code, and mathematical content. This dataset comprises approximately 300,000 instructions and covers a wide range of topics in each domain. Experiments show that our model could simultaneously achieve mastery of the three crucial domains.

♻ ☆ Navigating the EU AI Act: A Methodological Approach to Compliance for Safety-critical Products

J. Kelly, S. Zafar, L. Heidemann, J. Zacchi, D. Espinoza, N. Mata

In December 2023, the European Parliament provisionally agreed on the EU AI Act. This unprecedented regulatory framework for AI systems lays out guidelines to ensure the safety, legality, and trustworthiness of AI products. This paper presents a methodology for interpreting the EU AI Act requirements for high-risk AI systems by leveraging product quality models. We first propose an extended product quality model for AI systems, incorporating attributes relevant to the Act not covered by current quality models. We map the Act requirements to relevant quality attributes with the goal of refining them into measurable characteristics. We then propose a contract-based approach to derive technical requirements at the stakeholder level. This facilitates the development and assessment of AI systems that not only adhere to established quality standards, but also comply with the regulatory requirements outlined in the Act for high-risk (including safety-critical) AI systems. We demonstrate the applicability of this methodology on an exemplary automotive supply chain use case, where several stakeholders interact to achieve EU AI Act compliance.

comment: To be published in: 2024 IEEE Conference on Artificial Intelligence (CAI 2024)

♻ ☆ An Implicit GNN Solver for Poisson-like problems

Matthieu Nastorg, Michele Alessandro Bucci, Thibault Faney, Jean-Marc Gratien, Guillaume Charpiat, Marc Schoenauer

This paper presents $\Psi$-GNN, a novel Graph Neural Network (GNN) approach for solving the ubiquitous Poisson PDE problems with mixed boundary conditions. By leveraging the Implicit Layer Theory, $\Psi$-GNN models an "infinitely" deep network, thus avoiding the empirical tuning of the number of required Message Passing layers to attain the solution. Its original architecture explicitly takes into account the boundary conditions, a critical prerequisite for physical applications, and is able to adapt to any initially provided solution. $\Psi$-GNN is trained using a "physics-informed" loss, and the training process is stable by design, and insensitive to its initialization. Furthermore, the consistency of the approach is theoretically proven, and its flexibility and generalization efficiency are experimentally demonstrated: the same learned model can accurately handle unstructured meshes of various sizes, as well as different boundary conditions. To the best of our knowledge, $\Psi$-GNN is the first physics-informed GNN-based method that can handle various unstructured domains, boundary conditions and initial solutions while also providing convergence guarantees.

♻ ☆ Detecting Sexual Content at the Sentence Level in First Millennium Latin Texts

Thibault Clérice

In this study, we propose to evaluate the use of deep learning methods for semantic classification at the sentence level to accelerate the process of corpus building in the field of humanities and linguistics, a traditional and time-consuming task. We introduce a novel corpus comprising around 2500 sentences spanning from 300 BCE to 900 CE including sexual semantics (medical, erotica, etc.). We evaluate various sentence classification approaches and different input embedding layers, and show that all consistently outperform simple token-based searches. We explore the integration of idiolectal and sociolectal metadata embeddings (centuries, author, type of writing), but find that it leads to overfitting. Our results demonstrate the effectiveness of this approach, achieving high precision and true positive rates (TPR) of respectively 70.60% and 86.33% using HAN. We evaluate the impact of the dataset size on the model performances (420 instead of 2013), and show that, while our models perform worse, they still offer a high enough precision and TPR, even without MLM, respectively 69% and 51%. Given the result, we provide an analysis of the attention mechanism as a supporting added value for humanists in order to produce more data.

♻ ☆ Learning User Embeddings from Human Gaze for Personalised Saliency Prediction

Florian Strohm, Mihai Bâce, Andreas Bulling

Reusable embeddings of user behaviour have shown significant performance improvements for the personalised saliency prediction task. However, prior works require explicit user characteristics and preferences as input, which are often difficult to obtain. We present a novel method to extract user embeddings from pairs of natural images and corresponding saliency maps generated from a small amount of user-specific eye tracking data. At the core of our method is a Siamese convolutional neural encoder that learns the user embeddings by contrasting the image and personal saliency map pairs of different users. Evaluations on two public saliency datasets show that the generated embeddings have high discriminative power, are effective at refining universal saliency maps to the individual users, and generalise well across users and images. Finally, based on our model's ability to encode individual user characteristics, our work points towards other applications that can benefit from reusable embeddings of gaze behaviour.

♻ ☆ Born With a Silver Spoon? Investigating Socioeconomic Bias in Large Language Models

Smriti Singh, Shuvam Keshari, Vinija Jain, Aman Chadha

Socioeconomic bias in society exacerbates disparities, influencing access to opportunities and resources based on individuals' economic and social backgrounds. This pervasive issue perpetuates systemic inequalities, hindering the pursuit of inclusive progress as a society. In this paper, we investigate the presence of socioeconomic bias, if any, in large language models. To this end, we introduce a novel dataset SilverSpoon, consisting of 3000 samples that illustrate hypothetical scenarios that involve underprivileged people performing ethically ambiguous actions due to their circumstances, and ask whether the action is ethically justified. Further, this dataset has a dual-labeling scheme and has been annotated by people belonging to both ends of the socioeconomic spectrum. Using SilverSpoon, we evaluate the degree of socioeconomic bias expressed in large language models and the variation of this degree as a function of model size. We also perform qualitative analysis to analyze the nature of this bias. Our analysis reveals that while humans disagree on which situations require empathy toward the underprivileged, most large language models are unable to empathize with the socioeconomically underprivileged regardless of the situation. To foster further research in this domain, we make SilverSpoon and our evaluation harness publicly available.

♻ ☆ Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models

Joao Carvalho, An T. Le, Mark Baierl, Dorothea Koert, Jan Peters

Learning priors on trajectory distributions can help accelerate robot motion planning optimization. Given previously successful plans, learning trajectory generative models as priors for a new planning problem is highly desirable. Prior works propose several ways on utilizing this prior to bootstrapping the motion planning problem. Either sampling the prior for initializations or using the prior distribution in a maximum-a-posterior formulation for trajectory optimization. In this work, we propose learning diffusion models as priors. We then can sample directly from the posterior trajectory distribution conditioned on task goals, by leveraging the inverse denoising process of diffusion models. Furthermore, diffusion has been recently shown to effectively encode data multimodality in high-dimensional settings, which is particularly well-suited for large trajectory dataset. To demonstrate our method efficacy, we compare our proposed method - Motion Planning Diffusion - against several baselines in simulated planar robot and 7-dof robot arm manipulator environments. To assess the generalization capabilities of our method, we test it in environments with previously unseen obstacles. Our experiments show that diffusion models are strong priors to encode high-dimensional trajectory distributions of robot motions.

♻ ☆ Brain Networks and Intelligence: A Graph Neural Network Based Approach to Resting State fMRI Data

Bishal Thapaliya, Esra Akbas, Jiayu Chen, Raam Sapkota, Bhaskar Ray, Pranav Suresh, Vince Calhoun, Jingyu Liu

Resting-state functional magnetic resonance imaging (rsfMRI) is a powerful tool for investigating the relationship between brain function and cognitive processes as it allows for the functional organization of the brain to be captured without relying on a specific task or stimuli. In this paper, we present a novel modeling architecture called BrainRGIN for predicting intelligence (fluid, crystallized, and total intelligence) using graph neural networks on rsfMRI derived static functional network connectivity matrices. Extending from the existing graph convolution networks, our approach incorporates a clustering-based embedding and graph isomorphism network in the graph convolutional layer to reflect the nature of the brain sub-network organization and efficient network expression, in combination with TopK pooling and attention-based readout functions. We evaluated our proposed architecture on a large dataset, specifically the Adolescent Brain Cognitive Development Dataset, and demonstrated its effectiveness in predicting individual differences in intelligence. Our model achieved lower mean squared errors and higher correlation scores than existing relevant graph architectures and other traditional machine learning models for all of the intelligence prediction tasks. The middle frontal gyrus exhibited a significant contribution to both fluid and crystallized intelligence, suggesting their pivotal role in these cognitive processes. Total composite scores identified a diverse set of brain regions to be relevant which underscores the complex nature of total intelligence.

♻ ☆ NaNa and MiGu: Semantic Data Augmentation Techniques to Enhance Protein Classification in Graph Neural Networks

Yi-Shan Lan, Pin-Yu Chen, Tsung-Yi Ho

Protein classification tasks are essential in drug discovery. Real-world protein structures are dynamic, which will determine the properties of proteins. However, the existing machine learning methods, like ProNet (Wang et al., 2022a), only access limited conformational characteristics and protein side-chain features, leading to impractical protein structure and inaccuracy of protein classes in their predictions. In this paper, we propose novel semantic data augmentation methods, Novel Augmentation of New Node Attributes (NaNa), and Molecular Interactions and Geometric Upgrading (MiGu) to incorporate backbone chemical and side-chain biophysical information into protein classification tasks and a co-embedding residual learning framework. Specifically, we leverage molecular biophysical, secondary structure, chemical bonds, and ionic features of proteins to facilitate protein classification tasks. Furthermore, our semantic augmentation methods and the co-embedding residual learning framework can improve the performance of GIN (Xu et al., 2019) on EC and Fold datasets (Bairoch, 2000; Andreeva et al., 2007) by 16.41% and 11.33% respectively. Our code is available at https://github.com/r08b46009/Code_for_MIGU_NANA/tree/main.

♻ ☆ Graph Generation with $K^2$-trees ICLR

Yunhui Jang, Dongwoo Kim, Sungsoo Ahn

Generating graphs from a target distribution is a significant challenge across many domains, including drug discovery and social network analysis. In this work, we introduce a novel graph generation method leveraging $K^2$-tree representation, originally designed for lossless graph compression. The $K^2$-tree representation {encompasses inherent hierarchy while enabling compact graph generation}. In addition, we make contributions by (1) presenting a sequential $K^2$-treerepresentation that incorporates pruning, flattening, and tokenization processes and (2) introducing a Transformer-based architecture designed to generate the sequence by incorporating a specialized tree positional encoding scheme. Finally, we extensively evaluate our algorithm on four general and two molecular graph datasets to confirm its superiority for graph generation.

comment: International Conference on Learning Representations (ICLR) 2024

♻ ☆ A Simple and Scalable Representation for Graph Generation ICLR

Yunhui Jang, Seul Lee, Sungsoo Ahn

Recently, there has been a surge of interest in employing neural networks for graph generation, a fundamental statistical learning problem with critical applications like molecule design and community analysis. However, most approaches encounter significant limitations when generating large-scale graphs. This is due to their requirement to output the full adjacency matrices whose size grows quadratically with the number of nodes. In response to this challenge, we introduce a new, simple, and scalable graph representation named gap encoded edge list (GEEL) that has a small representation size that aligns with the number of edges. In addition, GEEL significantly reduces the vocabulary size by incorporating the gap encoding and bandwidth restriction schemes. GEEL can be autoregressively generated with the incorporation of node positional encoding, and we further extend GEEL to deal with attributed graphs by designing a new grammar. Our findings reveal that the adoption of this compact representation not only enhances scalability but also bolsters performance by simplifying the graph generation process. We conduct a comprehensive evaluation across ten non-attributed and two molecular graph generation tasks, demonstrating the effectiveness of GEEL.

comment: International Conference on Learning Representations (ICLR) 2024

♻ ☆ Take Care of Your Prompt Bias! Investigating and Mitigating Prompt Bias in Factual Knowledge Extraction COLING 2024

Ziyang Xu, Keqin Peng, Liang Ding, Dacheng Tao, Xiliang Lu

Recent research shows that pre-trained language models (PLMs) suffer from "prompt bias" in factual knowledge extraction, i.e., prompts tend to introduce biases toward specific labels. Prompt bias presents a significant challenge in assessing the factual knowledge within PLMs. Therefore, this paper aims to improve the reliability of existing benchmarks by thoroughly investigating and mitigating prompt bias. We show that: 1) all prompts in the experiments exhibit non-negligible bias, with gradient-based prompts like AutoPrompt and OptiPrompt displaying significantly higher levels of bias; 2) prompt bias can amplify benchmark accuracy unreasonably by overfitting the test datasets, especially on imbalanced datasets like LAMA. Based on these findings, we propose a representation-based approach to mitigate the prompt bias during inference time. Specifically, we first estimate the biased representation using prompt-only querying, and then remove it from the model's internal representations to generate the debiased representations, which are used to produce the final debiased outputs. Experiments across various prompts, PLMs, and benchmarks show that our approach can not only correct the overfitted performance caused by prompt bias, but also significantly improve the prompt retrieval capability (up to 10% absolute performance gain). These results indicate that our approach effectively alleviates prompt bias in knowledge evaluation, thereby enhancing the reliability of benchmark assessments. Hopefully, our plug-and-play approach can be a golden standard to strengthen PLMs toward reliable knowledge bases. Code and data are released in https://github.com/FelliYang/PromptBias.

comment: Accepted by COLING 2024

♻ ☆ Diffusion Models Generate Images Like Painters: an Analytical Theory of Outline First, Details Later NeurIPS23

Binxu Wang, John J. Vastola

How do diffusion generative models convert pure noise into meaningful images? In a variety of pretrained diffusion models (including conditional latent space models like Stable Diffusion), we observe that the reverse diffusion process that underlies image generation has the following properties: (i) individual trajectories tend to be low-dimensional and resemble 2D `rotations'; (ii) high-variance scene features like layout tend to emerge earlier, while low-variance details tend to emerge later; and (iii) early perturbations tend to have a greater impact on image content than later perturbations. To understand these phenomena, we derive and study a closed-form solution to the probability flow ODE for a Gaussian distribution, which shows that the reverse diffusion state rotates towards a gradually-specified target on the image manifold. It also shows that generation involves first committing to an outline, and then to finer and finer details. We find that this solution accurately describes the initial phase of image generation for pretrained models, and can in principle be used to make image generation more efficient by skipping reverse diffusion steps. Finally, we use our solution to characterize the image manifold in Stable Diffusion. Our viewpoint reveals an unexpected similarity between generation by GANs and diffusion and provides a conceptual link between diffusion and image retrieval.

comment: 44 pages, 28 figures. A briefer version was presented at NeurIPS23 Workshop on Diffusion Models [arXiv:2311.10892]

♻ ☆ Can Large Language Models Discern Evidence for Scientific Hypotheses? Case Studies in the Social Sciences

Sai Koneru, Jian Wu, Sarah Rajtmajer

Hypothesis formulation and testing are central to empirical research. A strong hypothesis is a best guess based on existing evidence and informed by a comprehensive view of relevant literature. However, with exponential increase in the number of scientific articles published annually, manual aggregation and synthesis of evidence related to a given hypothesis is a challenge. Our work explores the ability of current large language models (LLMs) to discern evidence in support or refute of specific hypotheses based on the text of scientific abstracts. We share a novel dataset for the task of scientific hypothesis evidencing using community-driven annotations of studies in the social sciences. We compare the performance of LLMs to several state-of-the-art benchmarks and highlight opportunities for future research in this area. The dataset is available at https://github.com/Sai90000/ScientificHypothesisEvidencing.git

♻ ☆ AIOS: LLM Agent Operating System

Kai Mei, Zelong Li, Shuyuan Xu, Ruosong Ye, Yingqiang Ge, Yongfeng Zhang

The integration and deployment of large language model (LLM)-based intelligent agents have been fraught with challenges that compromise their efficiency and efficacy. Among these issues are sub-optimal scheduling and resource allocation of agent requests over the LLM, the difficulties in maintaining context during interactions between agent and LLM, and the complexities inherent in integrating heterogeneous agents with different capabilities and specializations. The rapid increase of agent quantity and complexity further exacerbates these issues, often leading to bottlenecks and sub-optimal utilization of resources. Inspired by these challenges, this paper presents AIOS, an LLM agent operating system, which embeds large language model into operating systems (OS) as the brain of the OS, enabling an operating system "with soul" -- an important step towards AGI. Specifically, AIOS is designed to optimize resource allocation, facilitate context switch across agents, enable concurrent execution of agents, provide tool service for agents, and maintain access control for agents. We present the architecture of such an operating system, outline the core challenges it aims to resolve, and provide the basic design and implementation of the AIOS. Our experiments on concurrent execution of multiple agents demonstrate the reliability and efficiency of our AIOS modules. Through this, we aim to not only improve the performance and efficiency of LLM agents but also to pioneer for better development and deployment of the AIOS ecosystem in the future. The project is open-source at https://github.com/agiresearch/AIOS.

comment: 14 pages, 5 figures, 5 tables; comments and suggestions are appreciated

♻ ☆ Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators

Yinhong Liu, Han Zhou, Zhijiang Guo, Ehsan Shareghi, Ivan Vulić, Anna Korhonen, Nigel Collier

Large Language Models (LLMs) have demonstrated promising capabilities as automatic evaluators in assessing the quality of generated natural language. However, LLMs still exhibit biases in evaluation and often struggle to generate coherent evaluations that align with human assessments. In this work, we first conduct a systematic study of the misalignment between LLM evaluators and human judgement, revealing that existing calibration methods aimed at mitigating biases are insufficient for effectively aligning LLM evaluators. Inspired by the use of preference data in RLHF, we formulate the evaluation as a ranking problem and introduce Pairwise-preference Search (PairS), an uncertainty-guided search method that employs LLMs to conduct pairwise comparisons and efficiently ranks candidate texts. PairS achieves state-of-the-art performance on representative evaluation tasks and demonstrates significant improvements over direct scoring. Furthermore, we provide insights into the role of pairwise preference in quantifying the transitivity of LLMs and demonstrate how PairS benefits from calibration.

♻ ☆ Medical Speech Symptoms Classification via Disentangled Representation SC

Jianzong Wang, Pengcheng Li, Xulong Zhang, Ning Cheng, Jing Xiao

Intent is defined for understanding spoken language in existing works. Both textual features and acoustic features involved in medical speech contain intent, which is important for symptomatic diagnosis. In this paper, we propose a medical speech classification model named DRSC that automatically learns to disentangle intent and content representations from textual-acoustic data for classification. The intent representations of the text domain and the Mel-spectrogram domain are extracted via intent encoders, and then the reconstructed text feature and the Mel-spectrogram feature are obtained through two exchanges. After combining the intent from two domains into a joint representation, the integrated intent representation is fed into a decision layer for classification. Experimental results show that our model obtains an average accuracy rate of 95% in detecting 25 different medical symptoms.

comment: Accepted by the 27th International Conference on Computer Supported Cooperative Work in Design (CSCWD 2024)

♻ ☆ Learning Flexible Body Collision Dynamics with Hierarchical Contact Mesh Transformer ICLR 2024

Youn-Yeol Yu, Jeongwhan Choi, Woojin Cho, Kookjin Lee, Nayong Kim, Kiseok Chang, Chang-Seung Woo, Ilho Kim, Seok-Woo Lee, Joon-Young Yang, Sooyoung Yoon, Noseong Park

Recently, many mesh-based graph neural network (GNN) models have been proposed for modeling complex high-dimensional physical systems. Remarkable achievements have been made in significantly reducing the solving time compared to traditional numerical solvers. These methods are typically designed to i) reduce the computational cost in solving physical dynamics and/or ii) propose techniques to enhance the solution accuracy in fluid and rigid body dynamics. However, it remains under-explored whether they are effective in addressing the challenges of flexible body dynamics, where instantaneous collisions occur within a very short timeframe. In this paper, we present Hierarchical Contact Mesh Transformer (HCMT), which uses hierarchical mesh structures and can learn long-range dependencies (occurred by collisions) among spatially distant positions of a body -- two close positions in a higher-level mesh correspond to two distant positions in a lower-level mesh. HCMT enables long-range interactions, and the hierarchical mesh structure quickly propagates collision effects to faraway positions. To this end, it consists of a contact mesh Transformer and a hierarchical mesh Transformer (CMT and HMT, respectively). Lastly, we propose a flexible body dynamics dataset, consisting of trajectories that reflect experimental settings frequently used in the display industry for product designs. We also compare the performance of several baselines using well-known benchmark datasets. Our results show that HCMT provides significant performance improvements over existing methods. Our code is available at https://github.com/yuyudeep/hcmt.

comment: Accepted at ICLR 2024

♻ ☆ SelfIE: Self-Interpretation of Large Language Model Embeddings

Haozhe Chen, Carl Vondrick, Chengzhi Mao

How do large language models (LLMs) obtain their answers? The ability to explain and control an LLM's reasoning process is key for reliability, transparency, and future model developments. We propose SelfIE (Self-Interpretation of Embeddings), a framework that enables LLMs to interpret their own embeddings in natural language by leveraging their ability to respond to inquiries about a given passage. Capable of interpreting open-world concepts in the hidden embeddings, SelfIE reveals LLM internal reasoning in cases such as making ethical decisions, internalizing prompt injection, and recalling harmful knowledge. SelfIE's text descriptions on hidden embeddings also open up new avenues to control LLM reasoning. We propose Supervised Control, which allows editing open-ended concepts while only requiring gradient computation of individual layer. We extend RLHF to hidden embeddings and propose Reinforcement Control that erases harmful knowledge in LLM without supervision targets.

♻ ☆ Threats, Attacks, and Defenses in Machine Unlearning: A Survey

Ziyao Liu, Huanyi Ye, Chen Chen, Kwok-Yan Lam

Machine Unlearning (MU) has gained considerable attention recently for its potential to achieve Safe AI by removing the influence of specific data from trained machine learning models. This process, known as knowledge removal, addresses AI governance concerns of training data such as quality, sensitivity, copyright restrictions, and obsolescence. This capability is also crucial for ensuring compliance with privacy regulations such as the Right To Be Forgotten. Furthermore, effective knowledge removal mitigates the risk of harmful outcomes, safeguarding against biases, misinformation, and unauthorized data exploitation, thereby enhancing the safe and responsible use of AI systems. Efforts have been made to design efficient unlearning approaches, with MU services being examined for integration with existing machine learning as a service, allowing users to submit requests to remove specific data from the training corpus. However, recent research highlights vulnerabilities in machine unlearning systems, such as information leakage and malicious unlearning requests, that can lead to significant security and privacy concerns. Moreover, extensive research indicates that unlearning methods and prevalent attacks fulfill diverse roles within MU systems. For instance, unlearning can act as a mechanism to recover models from backdoor attacks, while backdoor attacks themselves can serve as an evaluation metric for unlearning effectiveness. This underscores the intricate relationship and complex interplay among these mechanisms in maintaining system functionality and safety. This survey aims to fill the gap between the extensive number of studies on threats, attacks, and defenses in machine unlearning and the absence of a comprehensive review that categorizes their taxonomy, methods, and solutions, thus offering valuable insights for future research directions and practical implementations.

♻ ☆ Near-Optimal Algorithms for Constrained k-Center Clustering with Instance-level Background Knowledge

Longkun Guo, Chaoqi Jia, Kewen Liao, Zhigang Lu, Minhui Xue

Center-based clustering has attracted significant research interest from both theory and practice. In many practical applications, input data often contain background knowledge that can be used to improve clustering results. In this work, we build on widely adopted $k$-center clustering and model its input background knowledge as must-link (ML) and cannot-link (CL) constraint sets. However, most clustering problems including $k$-center are inherently $\mathcal{NP}$-hard, while the more complex constrained variants are known to suffer severer approximation and computation barriers that significantly limit their applicability. By employing a suite of techniques including reverse dominating sets, linear programming (LP) integral polyhedron, and LP duality, we arrive at the first efficient approximation algorithm for constrained $k$-center with the best possible ratio of 2. We also construct competitive baseline algorithms and empirically evaluate our approximation algorithm against them on a variety of real datasets. The results validate our theoretical findings and demonstrate the great advantages of our algorithm in terms of clustering cost, clustering quality, and running time.

☆ SLEDGE: Synthesizing Simulation Environments for Driving Agents with Generative Models

Kashyap Chitta, Daniel Dauner, Andreas Geiger

SLEDGE is the first generative simulator for vehicle motion planning trained on real-world driving logs. Its core component is a learned model that is able to generate agent bounding boxes and lane graphs. The model's outputs serve as an initial state for traffic simulation. The unique properties of the entities to be generated for SLEDGE, such as their connectivity and variable count per scene, render the naive application of most modern generative models to this task non-trivial. Therefore, together with a systematic study of existing lane graph representations, we introduce a novel raster-to-vector autoencoder (RVAE). It encodes agents and the lane graph into distinct channels in a rasterized latent map. This facilitates both lane-conditioned agent generation and combined generation of lanes and agents with a Diffusion Transformer. Using generated entities in SLEDGE enables greater control over the simulation, e.g. upsampling turns or increasing traffic density. Further, SLEDGE can support 500m long routes, a capability not found in existing data-driven simulators like nuPlan. It presents new challenges for planning algorithms, evidenced by failure rates of over 40% for PDM, the winner of the 2023 nuPlan challenge, when tested on hard routes and dense traffic generated by our model. Compared to nuPlan, SLEDGE requires 500$\times$ less storage to set up (<4GB), making it a more accessible option and helping with democratizing future research in this field.

☆ The Need for Speed: Pruning Transformers with One Recipe ICLR

Samir Khaki, Konstantinos N. Plataniotis

We introduce the $\textbf{O}$ne-shot $\textbf{P}$runing $\textbf{T}$echnique for $\textbf{I}$nterchangeable $\textbf{N}$etworks ($\textbf{OPTIN}$) framework as a tool to increase the efficiency of pre-trained transformer architectures $\textit{without requiring re-training}$. Recent works have explored improving transformer efficiency, however often incur computationally expensive re-training procedures or depend on architecture-specific characteristics, thus impeding practical wide-scale adoption. To address these shortcomings, the OPTIN framework leverages intermediate feature distillation, capturing the long-range dependencies of model parameters (coined $\textit{trajectory}$), to produce state-of-the-art results on natural language, image classification, transfer learning, and semantic segmentation tasks $\textit{without re-training}$. Given a FLOP constraint, the OPTIN framework will compress the network while maintaining competitive accuracy performance and improved throughput. Particularly, we show a $\leq 2$% accuracy degradation from NLP baselines and a $0.5$% improvement from state-of-the-art methods on image classification at competitive FLOPs reductions. We further demonstrate the generalization of tasks and architecture with comparative performance using Mask2Former for semantic segmentation and cnn-style networks. OPTIN presents one of the first one-shot efficient frameworks for compressing transformer architectures that generalizes well across different class domains, in particular: natural language and image-related tasks, without $\textit{re-training}$.

comment: Accepted in the International Conference on Learning Representations (ICLR) 2024

☆ LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning

Rui Pan, Xiang Liu, Shizhe Diao, Renjie Pi, Jipeng Zhang, Chi Han, Tong Zhang

The machine learning community has witnessed impressive advancements since the first appearance of large language models (LLMs), yet their huge memory consumption has become a major roadblock to large-scale training. Parameter Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA) have been proposed to alleviate this problem, but their performance still fails to match full parameter training in most large-scale fine-tuning settings. Attempting to complement this deficiency, we investigate layerwise properties of LoRA on fine-tuning tasks and observe an uncommon skewness of weight norms across different layers. Utilizing this key observation, a surprisingly simple training strategy is discovered, which outperforms both LoRA and full parameter training in a wide range of settings with memory costs as low as LoRA. We name it Layerwise Importance Sampled AdamW (LISA), a promising alternative for LoRA, which applies the idea of importance sampling to different layers in LLMs and randomly freeze most middle layers during optimization. Experimental results show that with similar or less GPU memory consumption, LISA surpasses LoRA or even full parameter tuning in downstream fine-tuning tasks, where LISA consistently outperforms LoRA by over $11\%$-$37\%$ in terms of MT-Bench scores. On large models, specifically LLaMA-2-70B, LISA achieves on-par or better performance than LoRA on MT-Bench, GSM8K, and PubMedQA, demonstrating its effectiveness across different domains.

☆ CMP: Cooperative Motion Prediction with Multi-Agent Communication

Zhuoyuan Wu, Yuping Wang, Hengbo Ma, Zhaowei Li, Hang Qiu, Jiachen Li

The confluence of the advancement of Autonomous Vehicles (AVs) and the maturity of Vehicle-to-Everything (V2X) communication has enabled the capability of cooperative connected and automated vehicles (CAVs). Building on top of cooperative perception, this paper explores the feasibility and effectiveness of cooperative motion prediction. Our method, CMP, takes LiDAR signals as input to enhance tracking and prediction capabilities. Unlike previous work that focuses separately on either cooperative perception or motion prediction, our framework, to the best of our knowledge, is the first to address the unified problem where CAVs share information in both perception and prediction modules. Incorporated into our design is the unique capability to tolerate realistic V2X bandwidth limitations and transmission delays, while dealing with bulky perception representations. We also propose a prediction aggregation module, which unifies the predictions obtained by different CAVs and generates the final prediction. Through extensive experiments and ablation studies, we demonstrate the effectiveness of our method in cooperative perception, tracking, and motion prediction tasks. In particular, CMP reduces the average prediction error by 17.2\% with fewer missing detections compared with the no cooperation setting. Our work marks a significant step forward in the cooperative capabilities of CAVs, showcasing enhanced performance in complex scenarios.

☆ Serpent: Scalable and Efficient Image Restoration via Multi-scale Structured State Space Models

Mohammad Shahab Sepehri, Zalan Fabian, Mahdi Soltanolkotabi

The landscape of computational building blocks of efficient image restoration architectures is dominated by a combination of convolutional processing and various attention mechanisms. However, convolutional filters are inherently local and therefore struggle at modeling long-range dependencies in images. On the other hand, attention excels at capturing global interactions between arbitrary image regions, however at a quadratic cost in image dimension. In this work, we propose Serpent, an architecture that leverages recent advances in state space models (SSMs) in its core computational block. SSMs, originally introduced for sequence modeling, can maintain a global receptive field with a favorable linear scaling in input size. Our preliminary results demonstrate that Serpent can achieve reconstruction quality on par with state-of-the-art techniques, while requiring orders of magnitude less compute (up to $150$ fold reduction in FLOPS) and a factor of up to $5\times$ less GPU memory while maintaining a compact model size.

comment: 7 pages, 5 figures, preliminary workshop submission of a comprehensive work to be released soon

☆ Image-based Novel Fault Detection with Deep Learning Classifiers using Hierarchical Labels

Nurettin Sergin, Jiayu Huang, Tzyy-Shuh Chang, Hao Yan

One important characteristic of modern fault classification systems is the ability to flag the system when faced with previously unseen fault types. This work considers the unknown fault detection capabilities of deep neural network-based fault classifiers. Specifically, we propose a methodology on how, when available, labels regarding the fault taxonomy can be used to increase unknown fault detection performance without sacrificing model performance. To achieve this, we propose to utilize soft label techniques to improve the state-of-the-art deep novel fault detection techniques during the training process and novel hierarchically consistent detection statistics for online novel fault detection. Finally, we demonstrated increased detection performance on novel fault detection in inspection images from the hot steel rolling process, with results well replicated across multiple scenarios and baseline detection methods.

comment: Accepted in IISE Transaction

☆ Large scale paired antibody language models

Henry Kenlay, Frédéric A. Dreyer, Aleksandr Kovaltsuk, Dom Miketa, Douglas Pires, Charlotte M. Deane

Antibodies are proteins produced by the immune system that can identify and neutralise a wide variety of antigens with high specificity and affinity, and constitute the most successful class of biotherapeutics. With the advent of next-generation sequencing, billions of antibody sequences have been collected in recent years, though their application in the design of better therapeutics has been constrained by the sheer volume and complexity of the data. To address this challenge, we present IgBert and IgT5, the best performing antibody-specific language models developed to date which can consistently handle both paired and unpaired variable region sequences as input. These models are trained comprehensively using the more than two billion unpaired sequences and two million paired sequences of light and heavy chains present in the Observed Antibody Space dataset. We show that our models outperform existing antibody and protein language models on a diverse range of design and regression tasks relevant to antibody engineering. This advancement marks a significant leap forward in leveraging machine learning, large scale data sets and high-performance computing for enhancing antibody design for therapeutic development.

comment: 14 pages, 2 figures, 6 tables, model weights available at https://zenodo.org/doi/10.5281/zenodo.10876908

☆ The Unreasonable Ineffectiveness of the Deeper Layers

Andrey Gromov, Kushal Tirumala, Hassan Shapourian, Paolo Glorioso, Daniel A. Roberts

We empirically study a simple layer-pruning strategy for popular families of open-weight pretrained LLMs, finding minimal degradation of performance on different question-answering benchmarks until after a large fraction (up to half) of the layers are removed. To prune these models, we identify the optimal block of layers to prune by considering similarity across layers; then, to "heal" the damage, we perform a small amount of finetuning. In particular, we use parameter-efficient finetuning (PEFT) methods, specifically quantization and Low Rank Adapters (QLoRA), such that each of our experiments can be performed on a single A100 GPU. From a practical perspective, these results suggest that layer pruning methods can complement other PEFT strategies to further reduce computational resources of finetuning on the one hand, and can improve the memory and latency of inference on the other hand. From a scientific perspective, the robustness of these LLMs to the deletion of layers implies either that current pretraining methods are not properly leveraging the parameters in the deeper layers of the network or that the shallow layers play a critical role in storing knowledge.

comment: 12 + 10 pages, 5 + 4 figures

☆ Compressed Multi-task embeddings for Data-Efficient Downstream training and inference in Earth Observation RSS 2024

Carlos Gomes, Thomas Brunschwiler

As repositories of large scale data in earth observation (EO) have grown, so have transfer and storage costs for model training and inference, expending significant resources. We introduce Neural Embedding Compression (NEC), based on the transfer of compressed embeddings to data consumers instead of raw data. We adapt foundation models (FM) through learned neural compression to generate multi-task embeddings while navigating the tradeoff between compression rate and embedding utility. We update only a small fraction of the FM parameters (10%) for a short training period (1% of the iterations of pre-training). We evaluate NEC on two EO tasks: scene classification and semantic segmentation. Compared with applying traditional compression to the raw data, NEC achieves similar accuracy with a 75% to 90% reduction in data. Even at 99.7% compression, performance drops by only 5% on the scene classification task. Overall, NEC is a data-efficient yet performant approach for multi-task EO modelling.

comment: Published at IGARSS 2024

☆ Empowering Data Mesh with Federated Learning KDD

Haoyuan Li, Salman Toor

The evolution of data architecture has seen the rise of data lakes, aiming to solve the bottlenecks of data management and promote intelligent decision-making. However, this centralized architecture is limited by the proliferation of data sources and the growing demand for timely analysis and processing. A new data paradigm, Data Mesh, is proposed to overcome these challenges. Data Mesh treats domains as a first-class concern by distributing the data ownership from the central team to each data domain, while keeping the federated governance to monitor domains and their data products. Many multi-million dollar organizations like Paypal, Netflix, and Zalando have already transformed their data analysis pipelines based on this new architecture. In this decentralized architecture where data is locally preserved by each domain team, traditional centralized machine learning is incapable of conducting effective analysis across multiple domains, especially for security-sensitive organizations. To this end, we introduce a pioneering approach that incorporates Federated Learning into Data Mesh. To the best of our knowledge, this is the first open-source applied work that represents a critical advancement toward the integration of federated learning methods into the Data Mesh paradigm, underscoring the promising prospects for privacy-preserving and decentralized data analysis strategies within Data Mesh architecture.

comment: In Proceedings of ACM Knowledge Discovery and Data Mining, Barcelona, Spain, 25th - 29th August, 2024 (Conference acronym KDD), 9 pages

☆ Sample complexity of quantum hypothesis testing

Hao-Chung Cheng, Nilanjana Datta, Nana Liu, Theshani Nuradha, Robert Salzmann, Mark M. Wilde

Quantum hypothesis testing has been traditionally studied from the information-theoretic perspective, wherein one is interested in the optimal decay rate of error probabilities as a function of the number of samples of an unknown state. In this paper, we study the sample complexity of quantum hypothesis testing, wherein the goal is to determine the minimum number of samples needed to reach a desired error probability. By making use of the wealth of knowledge that already exists in the literature on quantum hypothesis testing, we characterize the sample complexity of binary quantum hypothesis testing in the symmetric and asymmetric settings, and we provide bounds on the sample complexity of multiple quantum hypothesis testing. In more detail, we prove that the sample complexity of symmetric binary quantum hypothesis testing depends logarithmically on the inverse error probability and inversely on the negative logarithm of the fidelity. As a counterpart of the quantum Stein's lemma, we also find that the sample complexity of asymmetric binary quantum hypothesis testing depends logarithmically on the inverse type~II error probability and inversely on the quantum relative entropy. Finally, we provide lower and upper bounds on the sample complexity of multiple quantum hypothesis testing, with it remaining an intriguing open question to improve these bounds.

comment: 38 pages, 1 figure, preliminary version; see independent and concurrent work of Pensia, Jog, Loh at arXiv:2403.16981

☆ Using Domain Knowledge to Guide Dialog Structure Induction via Neural Probabilistic Soft Logic

Connor Pryor, Quan Yuan, Jeremiah Liu, Mehran Kazemi, Deepak Ramachandran, Tania Bedrax-Weiss, Lise Getoor

Dialog Structure Induction (DSI) is the task of inferring the latent dialog structure (i.e., a set of dialog states and their temporal transitions) of a given goal-oriented dialog. It is a critical component for modern dialog system design and discourse analysis. Existing DSI approaches are often purely data-driven, deploy models that infer latent states without access to domain knowledge, underperform when the training corpus is limited/noisy, or have difficulty when test dialogs exhibit distributional shifts from the training domain. This work explores a neural-symbolic approach as a potential solution to these problems. We introduce Neural Probabilistic Soft Logic Dialogue Structure Induction (NEUPSL DSI), a principled approach that injects symbolic knowledge into the latent space of a generative neural model. We conduct a thorough empirical investigation on the effect of NEUPSL DSI learning on hidden representation quality, few-shot learning, and out-of-domain generalization performance. Over three dialog structure induction datasets and across unsupervised and semi-supervised settings for standard and cross-domain generalization, the injection of symbolic knowledge using NEUPSL DSI provides a consistent boost in performance over the canonical baselines.

☆ Counterfactual Fairness through Transforming Data Orthogonal to Bias

Shuyi Chen, Shixiang Zhu

Machine learning models have shown exceptional prowess in solving complex issues across various domains. Nonetheless, these models can sometimes exhibit biased decision-making, leading to disparities in treatment across different groups. Despite the extensive research on fairness, the nuanced effects of multivariate and continuous sensitive variables on decision-making outcomes remain insufficiently studied. We introduce a novel data pre-processing algorithm, Orthogonal to Bias (OB), designed to remove the influence of a group of continuous sensitive variables, thereby facilitating counterfactual fairness in machine learning applications. Our approach is grounded in the assumption of a jointly normal distribution within a structural causal model (SCM), proving that counterfactual fairness can be achieved by ensuring the data is uncorrelated with sensitive variables. The OB algorithm is model-agnostic, catering to a wide array of machine learning models and tasks, and includes a sparse variant to enhance numerical stability through regularization. Through empirical evaluation on simulated and real-world datasets - including the adult income and the COMPAS recidivism datasets - our methodology demonstrates its capacity to enable fairer outcomes without compromising accuracy.

☆ Climate Downscaling: A Deep-Learning Based Super-resolution Model of Precipitation Data with Attention Block and Skip Connections

Chia-Hao Chiang, Zheng-Han Huang, Liwen Liu, Hsin-Chien Liang, Yi-Chi Wang, Wan-Ling Tseng, Chao Wang, Che-Ta Chen, Ko-Chih Wang

Human activities accelerate consumption of fossil fuels and produce greenhouse gases, resulting in urgent issues today: global warming and the climate change. These indirectly cause severe natural disasters, plenty of lives suffering and huge losses of agricultural properties. To mitigate impacts on our lands, scientists are developing renewable, reusable, and clean energies and climatologists are trying to predict the extremes. Meanwhile, governments are publicizing resource-saving policies for a more eco-friendly society and arousing environment awareness. One of the most influencing factors is the precipitation, bringing condensed water vapor onto lands. Water resources are the most significant but basic needs in society, not only supporting our livings, but also economics. In Taiwan, although the average annual precipitation is up to 2,500 millimeter (mm), the water allocation for each person is lower than the global average due to drastically geographical elevation changes and uneven distribution through the year. Thus, it is crucial to track and predict the rainfall to make the most use of it and to prevent the floods. However, climate models have limited resolution and require intensive computational power for local-scale use. Therefore, we proposed a deep convolutional neural network with skip connections, attention blocks, and auxiliary data concatenation, in order to downscale the low-resolution precipitation data into high-resolution one. Eventually, we compare with other climate downscaling methods and show better performance in metrics of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Pearson Correlation, structural similarity index (SSIM), and forecast indicators.

★ Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation

Abdelrhman Werby, Chenguang Huang, Martin Büchner, Abhinav Valada, Wolfram Burgard

Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features. While these maps allow for the prediction of point-wise saliency maps when queried for a certain language concept, large-scale environments and abstract queries beyond the object level still pose a considerable hurdle, ultimately limiting language-grounded robotic navigation. In this work, we present HOV-SG, a hierarchical open-vocabulary 3D scene graph mapping approach for language-grounded robot navigation. Leveraging open-vocabulary vision foundation models, we first obtain state-of-the-art open-vocabulary segment-level maps in 3D and subsequently construct a 3D scene graph hierarchy consisting of floor, room, and object concepts, each enriched with open-vocabulary features. Our approach is able to represent multi-story buildings and allows robotic traversal of those using a cross-floor Voronoi graph. HOV-SG is evaluated on three distinct datasets and surpasses previous baselines in open-vocabulary semantic accuracy on the object, room, and floor level while producing a 75% reduction in representation size compared to dense open-vocabulary maps. In order to prove the efficacy and generalization capabilities of HOV-SG, we showcase successful long-horizon language-conditioned robot navigation within real-world multi-storage environments. We provide code and trial video data at http://hovsg.github.io/.

comment: Code and video are available at http://hovsg.github.io/

☆ TractOracle: towards an anatomically-informed reward function for RL-based tractography

Antoine Théberge, Maxime Descoteaux, Pierre-Marc Jodoin

Reinforcement learning (RL)-based tractography is a competitive alternative to machine learning and classical tractography algorithms due to its high anatomical accuracy obtained without the need for any annotated data. However, the reward functions so far used to train RL agents do not encapsulate anatomical knowledge which causes agents to generate spurious false positives tracts. In this paper, we propose a new RL tractography system, TractOracle, which relies on a reward network trained for streamline classification. This network is used both as a reward function during training as well as a mean for stopping the tracking process early and thus reduce the number of false positive streamlines. This makes our system a unique method that evaluates and reconstructs WM streamlines at the same time. We report an improvement of true positive ratios by almost 20\% and a reduction of 3x of false positive ratios on one dataset and an increase between 2x and 7x in the number true positive streamlines on another dataset.

☆ Mechanistic Design and Scaling of Hybrid Architectures

Michael Poli, Armin W Thomas, Eric Nguyen, Pragaash Ponnusamy, Björn Deiseroth, Kristian Kersting, Taiji Suzuki, Brian Hie, Stefano Ermon, Christopher Ré, Ce Zhang, Stefano Massaroli

The development of deep learning architectures is a resource-demanding process, due to a vast design space, long prototyping times, and high compute costs associated with at-scale model training and evaluation. We set out to simplify this process by grounding it in an end-to-end mechanistic architecture design (MAD) pipeline, encompassing small-scale capability unit tests predictive of scaling laws. Through a suite of synthetic token manipulation tasks such as compression and recall, designed to probe capabilities, we identify and test new hybrid architectures constructed from a variety of computational primitives. We experimentally validate the resulting architectures via an extensive compute-optimal and a new state-optimal scaling law analysis, training over 500 language models between 70M to 7B parameters. Surprisingly, we find MAD synthetics to correlate with compute-optimal perplexity, enabling accurate evaluation of new architectures via isolated proxy tasks. The new architectures found via MAD, based on simple ideas such as hybridization and sparsity, outperform state-of-the-art Transformer, convolutional, and recurrent architectures (Transformer++, Hyena, Mamba) in scaling, both at compute-optimal budgets and in overtrained regimes. Overall, these results provide evidence that performance on curated synthetic tasks can be predictive of scaling laws, and that an optimal architecture should leverage specialized layers via a hybrid topology.

☆ GTA-HDR: A Large-Scale Synthetic Dataset for HDR Image Reconstruction

Hrishav Bakul Barua, Kalin Stefanov, KokSheik Wong, Abhinav Dhall, Ganesh Krishnasamy

High Dynamic Range (HDR) content (i.e., images and videos) has a broad range of applications. However, capturing HDR content from real-world scenes is expensive and time- consuming. Therefore, the challenging task of reconstructing visually accurate HDR images from their Low Dynamic Range (LDR) counterparts is gaining attention in the vision research community. A major challenge in this research problem is the lack of datasets, which capture diverse scene conditions (e.g., lighting, shadows, weather, locations, landscapes, objects, humans, buildings) and various image features (e.g., color, contrast, saturation, hue, luminance, brightness, radiance). To address this gap, in this paper, we introduce GTA-HDR, a large-scale synthetic dataset of photo-realistic HDR images sampled from the GTA-V video game. We perform thorough evaluation of the proposed dataset, which demonstrates significant qualitative and quantitative improvements of the state-of-the-art HDR image reconstruction methods. Furthermore, we demonstrate the effectiveness of the proposed dataset and its impact on additional computer vision tasks including 3D human pose estimation, human body part segmentation, and holistic scene segmentation. The dataset, data collection pipeline, and evaluation code are available at: https://github.com/HrishavBakulBarua/GTA-HDR.

comment: Submitted to IEEE

☆ GPFL: A Gradient Projection-Based Client Selection Framework for Efficient Federated Learning

Shijie Na, Yuzhi Liang, Siu-Ming Yiu

Federated learning client selection is crucial for determining participant clients while balancing model accuracy and communication efficiency. Existing methods have limitations in handling data heterogeneity, computational burdens, and independent client treatment. To address these challenges, we propose GPFL, which measures client value by comparing local and global descent directions. We also employ an Exploit-Explore mechanism to enhance performance. Experimental results on FEMINST and CIFAR-10 datasets demonstrate that GPFL outperforms baselines in Non-IID scenarios, achieving over 9\% improvement in FEMINST test accuracy. Moreover, GPFL exhibits shorter computation times through pre-selection and parameter reuse in federated learning.

comment: 8 pages, 5 figures

☆ Learning the Optimal Power Flow: Environment Design Matters

Thomas Wolgast, Astrid Nieße

To solve the optimal power flow (OPF) problem, reinforcement learning (RL) emerges as a promising new approach. However, the RL-OPF literature is strongly divided regarding the exact formulation of the OPF problem as an RL environment. In this work, we collect and implement diverse environment design decisions from the literature regarding training data, observation space, episode definition, and reward function choice. In an experimental analysis, we show the significant impact of these environment design options on RL-OPF training performance. Further, we derive some first recommendations regarding the choice of these design decisions. The created environment framework is fully open-source and can serve as a benchmark for future research in the RL-OPF field.

☆ DiffH2O: Diffusion-Based Synthesis of Hand-Object Interactions from Textual Descriptions

Sammy Christen, Shreyas Hampali, Fadime Sener, Edoardo Remelli, Tomas Hodan, Eric Sauser, Shugao Ma, Bugra Tekin

Generating natural hand-object interactions in 3D is challenging as the resulting hand and object motions are expected to be physically plausible and semantically meaningful. Furthermore, generalization to unseen objects is hindered by the limited scale of available hand-object interaction datasets. We propose DiffH2O, a novel method to synthesize realistic, one or two-handed object interactions from provided text prompts and geometry of the object. The method introduces three techniques that enable effective learning from limited data. First, we decompose the task into a grasping stage and a text-based interaction stage and use separate diffusion models for each. In the grasping stage, the model only generates hand motions, whereas in the interaction phase both hand and object poses are synthesized. Second, we propose a compact representation that tightly couples hand and object poses. Third, we propose two different guidance schemes to allow more control of the generated motions: grasp guidance and detailed textual guidance. Grasp guidance takes a single target grasping pose and guides the diffusion model to reach this grasp at the end of the grasping stage, which provides control over the grasping pose. Given a grasping motion from this stage, multiple different actions can be prompted in the interaction phase. For textual guidance, we contribute comprehensive text descriptions to the GRAB dataset and show that they enable our method to have more fine-grained control over hand-object interactions. Our quantitative and qualitative evaluation demonstrates that the proposed method outperforms baseline methods and leads to natural hand-object motions. Moreover, we demonstrate the practicality of our framework by utilizing a hand pose estimate from an off-the-shelf pose estimator for guidance, and then sampling multiple different actions in the interaction stage.

comment: Project Page: https://diffh2o.github.io/

☆ Are Compressed Language Models Less Subgroup Robust? EMNLP 2023

Leonidas Gee, Andrea Zugarini, Novi Quadrianto

To reduce the inference cost of large language models, model compression is increasingly used to create smaller scalable models. However, little is known about their robustness to minority subgroups defined by the labels and attributes of a dataset. In this paper, we investigate the effects of 18 different compression methods and settings on the subgroup robustness of BERT language models. We show that worst-group performance does not depend on model size alone, but also on the compression method used. Additionally, we find that model compression does not always worsen the performance on minority subgroups. Altogether, our analysis serves to further research into the subgroup robustness of model compression.

comment: The 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023)

☆ Annotated Biomedical Video Generation using Denoising Diffusion Probabilistic Models and Flow Fields

Rüveyda Yilmaz, Dennis Eschweiler, Johannes Stegmaier

The segmentation and tracking of living cells play a vital role within the biomedical domain, particularly in cancer research, drug development, and developmental biology. These are usually tedious and time-consuming tasks that are traditionally done by biomedical experts. Recently, to automatize these processes, deep learning based segmentation and tracking methods have been proposed. These methods require large-scale datasets and their full potential is constrained by the scarcity of annotated data in the biomedical imaging domain. To address this limitation, we propose Biomedical Video Diffusion Model (BVDM), capable of generating realistic-looking synthetic microscopy videos. Trained only on a single real video, BVDM can generate videos of arbitrary length with pixel-level annotations that can be used for training data-hungry models. It is composed of a denoising diffusion probabilistic model (DDPM) generating high-fidelity synthetic cell microscopy images and a flow prediction model (FPM) predicting the non-rigid transformation between consecutive video frames. During inference, initially, the DDPM imposes realistic cell textures on synthetic cell masks which are generated based on real data statistics. The flow prediction model predicts the flow field between consecutive masks and applies that to the DDPM output from the previous time frame to create the next one while keeping temporal consistency. BVDM outperforms state-of-the-art synthetic live cell microscopy video generation models. Furthermore, we demonstrate that a sufficiently large synthetic dataset enhances the performance of cell segmentation and tracking models compared to using a limited amount of available real data.

☆ Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms

Michael Hanna, Sandro Pezzelle, Yonatan Belinkov

Many recent language model (LM) interpretability studies have adopted the circuits framework, which aims to find the minimal computational subgraph, or circuit, that explains LM behavior on a given task. Most studies determine which edges belong in a LM's circuit by performing causal interventions on each edge independently, but this scales poorly with model size. Edge attribution patching (EAP), gradient-based approximation to interventions, has emerged as a scalable but imperfect solution to this problem. In this paper, we introduce a new method - EAP with integrated gradients (EAP-IG) - that aims to better maintain a core property of circuits: faithfulness. A circuit is faithful if all model edges outside the circuit can be ablated without changing the model's performance on the task; faithfulness is what justifies studying circuits, rather than the full model. Our experiments demonstrate that circuits found using EAP are less faithful than those found using EAP-IG, even though both have high node overlap with circuits found previously using causal interventions. We conclude more generally that when using circuits to compare the mechanisms models use to solve tasks, faithfulness, not overlap, is what should be measured.

☆ Scenario-Based Curriculum Generation for Multi-Agent Autonomous Driving

Axel Brunnbauer, Luigi Berducci, Peter Priller, Dejan Nickovic, Radu Grosu

The automated generation of diverse and complex training scenarios has been an important ingredient in many complex learning tasks. Especially in real-world application domains, such as autonomous driving, auto-curriculum generation is considered vital for obtaining robust and general policies. However, crafting traffic scenarios with multiple, heterogeneous agents is typically considered as a tedious and time-consuming task, especially in more complex simulation environments. In our work, we introduce MATS-Gym, a Multi-Agent Traffic Scenario framework to train agents in CARLA, a high-fidelity driving simulator. MATS-Gym is a multi-agent training framework for autonomous driving that uses partial scenario specifications to generate traffic scenarios with variable numbers of agents. This paper unifies various existing approaches to traffic scenario description into a single training framework and demonstrates how it can be integrated with techniques from unsupervised environment design to automate the generation of adaptive auto-curricula. The code is available at https://github.com/AutonomousDrivingExaminer/mats-gym.

comment: 7 Pages, Under Review

☆ Secure Aggregation is Not Private Against Membership Inference Attacks

Khac-Hoang Ngo, Johan Östman, Giuseppe Durisi, Alexandre Graell i Amat

Secure aggregation (SecAgg) is a commonly-used privacy-enhancing mechanism in federated learning, affording the server access only to the aggregate of model updates while safeguarding the confidentiality of individual updates. Despite widespread claims regarding SecAgg's privacy-preserving capabilities, a formal analysis of its privacy is lacking, making such presumptions unjustified. In this paper, we delve into the privacy implications of SecAgg by treating it as a local differential privacy (LDP) mechanism for each local update. We design a simple attack wherein an adversarial server seeks to discern which update vector a client submitted, out of two possible ones, in a single training round of federated learning under SecAgg. By conducting privacy auditing, we assess the success probability of this attack and quantify the LDP guarantees provided by SecAgg. Our numerical results unveil that, contrary to prevailing claims, SecAgg offers weak privacy against membership inference attacks even in a single training round. Indeed, it is difficult to hide a local update by adding other independent local updates when the updates are of high dimension. Our findings underscore the imperative for additional privacy-enhancing mechanisms, such as noise injection, in federated learning.

☆ SciNews: From Scholarly Complexities to Public Narratives -- A Dataset for Scientific News Report Generation LREC

Dongqi Pu, Yifan Wang, Jia Loy, Vera Demberg

Scientific news reports serve as a bridge, adeptly translating complex research articles into reports that resonate with the broader public. The automated generation of such narratives enhances the accessibility of scholarly insights. In this paper, we present a new corpus to facilitate this paradigm development. Our corpus comprises a parallel compilation of academic publications and their corresponding scientific news reports across nine disciplines. To demonstrate the utility and reliability of our dataset, we conduct an extensive analysis, highlighting the divergences in readability and brevity between scientific news narratives and academic manuscripts. We benchmark our dataset employing state-of-the-art text generation models. The evaluation process involves both automatic and human evaluation, which lays the groundwork for future explorations into the automated generation of scientific news reports. The dataset and code related to this work are available at https://dongqi.me/projects/SciNews.

comment: LREC-COLING 2024 Main Conference Paper

☆ Noise2Noise Denoising of CRISM Hyperspectral Data ICLR 2024

Robert Platt, Rossella Arcucci, Cédric M. John

Hyperspectral data acquired by the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) have allowed for unparalleled mapping of the surface mineralogy of Mars. Due to sensor degradation over time, a significant portion of the recently acquired data is considered unusable. Here a new data-driven model architecture, Noise2Noise4Mars (N2N4M), is introduced to remove noise from CRISM images. Our model is self-supervised and does not require zero-noise target data, making it well suited for use in Planetary Science applications where high quality labelled data is scarce. We demonstrate its strong performance on synthetic-noise data and CRISM images, and its impact on downstream classification performance, outperforming benchmark methods on most metrics. This allows for detailed analysis for critical sites of interest on the Martian surface, including proposed lander sites.

comment: 5 pages, 3 figures. Accepted as a conference paper at the ICLR 2024 ML4RS Workshop

☆ CCDSReFormer: Traffic Flow Prediction with a Criss-Crossed Dual-Stream Enhanced Rectified Transformer Model

Zhiqi Shao, Michael G. H. Bell, Ze Wang, D. Glenn Geers, Xusheng Yao, Junbin Gao

Accurate, and effective traffic forecasting is vital for smart traffic systems, crucial in urban traffic planning and management. Current Spatio-Temporal Transformer models, despite their prediction capabilities, struggle with balancing computational efficiency and accuracy, favoring global over local information, and handling spatial and temporal data separately, limiting insight into complex interactions. We introduce the Criss-Crossed Dual-Stream Enhanced Rectified Transformer model (CCDSReFormer), which includes three innovative modules: Enhanced Rectified Spatial Self-attention (ReSSA), Enhanced Rectified Delay Aware Self-attention (ReDASA), and Enhanced Rectified Temporal Self-attention (ReTSA). These modules aim to lower computational needs via sparse attention, focus on local information for better traffic dynamics understanding, and merge spatial and temporal insights through a unique learning method. Extensive tests on six real-world datasets highlight CCDSReFormer's superior performance. An ablation study also confirms the significant impact of each component on the model's predictive accuracy, showcasing our model's ability to forecast traffic flow effectively.

comment: 18 pages

☆ Leave No Patient Behind: Enhancing Medication Recommendation for Rare Disease Patients

Zihao Zhao, Yi Jing, Fuli Feng, Jiancan Wu, Chongming Gao, Xiangnan He

Medication recommendation systems have gained significant attention in healthcare as a means of providing tailored and effective drug combinations based on patients' clinical information. However, existing approaches often suffer from fairness issues, as recommendations tend to be more accurate for patients with common diseases compared to those with rare conditions. In this paper, we propose a novel model called Robust and Accurate REcommendations for Medication (RAREMed), which leverages the pretrain-finetune learning paradigm to enhance accuracy for rare diseases. RAREMed employs a transformer encoder with a unified input sequence approach to capture complex relationships among disease and procedure codes. Additionally, it introduces two self-supervised pre-training tasks, namely Sequence Matching Prediction (SMP) and Self Reconstruction (SR), to learn specialized medication needs and interrelations among clinical codes. Experimental results on two real-world datasets demonstrate that RAREMed provides accurate drug sets for both rare and common disease patients, thereby mitigating unfairness in medication recommendation systems.

☆ EulerFormer: Sequential User Behavior Modeling with Complex Vector Attention SIGIR'24

Zhen Tian, Wayne Xin Zhao, Changwang Zhang, Xin Zhao, Zhongrui Ma, Ji-Rong Wen

To capture user preference, transformer models have been widely applied to model sequential user behavior data. The core of transformer architecture lies in the self-attention mechanism, which computes the pairwise attention scores in a sequence. Due to the permutation-equivariant nature, positional encoding is used to enhance the attention between token representations. In this setting, the pairwise attention scores can be derived by both semantic difference and positional difference. However, prior studies often model the two kinds of difference measurements in different ways, which potentially limits the expressive capacity of sequence modeling. To address this issue, this paper proposes a novel transformer variant with complex vector attention, named EulerFormer, which provides a unified theoretical framework to formulate both semantic difference and positional difference. The EulerFormer involves two key technical improvements. First, it employs a new transformation function for efficiently transforming the sequence tokens into polar-form complex vectors using Euler's formula, enabling the unified modeling of both semantic and positional information in a complex rotation form.Secondly, it develops a differential rotation mechanism, where the semantic rotation angles can be controlled by an adaptation function, enabling the adaptive integration of the semantic and positional information according to the semantic contexts.Furthermore, a phase contrastive learning task is proposed to improve the anisotropy of contextual representations in EulerFormer. Our theoretical framework possesses a high degree of completeness and generality. It is more robust to semantic variations and possesses moresuperior theoretical properties in principle. Extensive experiments conducted on four public datasets demonstrate the effectiveness and efficiency of our approach.

comment: Accepted for publication in SIGIR'24

☆ Masked Autoencoders are PDE Learners

Anthony Zhou, Amir Barati Farimani

Neural solvers for partial differential equations (PDEs) have great potential, yet their practicality is currently limited by their generalizability. PDEs evolve over broad scales and exhibit diverse behaviors; predicting these phenomena will require learning representations across a wide variety of inputs, which may encompass different coefficients, geometries, or equations. As a step towards generalizable PDE modeling, we adapt masked pretraining for PDEs. Through self-supervised learning across PDEs, masked autoencoders can learn useful latent representations for downstream tasks. In particular, masked pretraining can improve coefficient regression and timestepping performance of neural solvers on unseen equations. We hope that masked pretraining can emerge as a unifying method across large, unlabeled, and heterogeneous datasets to learn latent physics at scale.

comment: 10 pages, 3 figures

☆ Rotate to Scan: UNet-like Mamba with Triplet SSM Module for Medical Image Segmentation

Hao Tang, Lianglun Cheng, Guoheng Huang, Zhengguang Tan, Junhao Lu, Kaihong Wu

Image segmentation holds a vital position in the realms of diagnosis and treatment within the medical domain. Traditional convolutional neural networks (CNNs) and Transformer models have made significant advancements in this realm, but they still encounter challenges because of limited receptive field or high computing complexity. Recently, State Space Models (SSMs), particularly Mamba and its variants, have demonstrated notable performance in the field of vision. However, their feature extraction methods may not be sufficiently effective and retain some redundant structures, leaving room for parameter reduction. Motivated by previous spatial and channel attention methods, we propose Triplet Mamba-UNet. The method leverages residual VSS Blocks to extract intensive contextual features, while Triplet SSM is employed to fuse features across spatial and channel dimensions. We conducted experiments on ISIC17, ISIC18, CVC-300, CVC-ClinicDB, Kvasir-SEG, CVC-ColonDB, and Kvasir-Instrument datasets, demonstrating the superior segmentation performance of our proposed TM-UNet. Additionally, compared to the previous VM-UNet, our model achieves a one-third reduction in parameters.

☆ MEP: Multiple Kernel Learning Enhancing Relative Positional Encoding Length Extrapolation

Weiguo Gao

When the predicted sequence length exceeds the length seen during training, the transformer's inference accuracy diminishes. Existing relative position encoding methods, such as those based on the ALiBi technique, address the length extrapolation challenge exclusively through the implementation of a single kernel function, which introduces a constant bias to every post-softmax attention scores according to their distance. These approaches do not investigate or employ multiple kernel functions to address the extrapolation challenge. Drawing on the ALiBi approach, this study proposes a novel relative positional encoding method, called MEP, which employs a weighted average to combine distinct kernel functions(such as the exponential kernel and the Gaussian kernel) to generate a bias that is applied to post-softmax attention scores. Initially, the framework utilizes various kernel functions to construct multiple kernel functions. Each kernel function adheres to a consistent mean weight coefficient, harnessing the synergistic advantages of different kernels to formulate an innovative bias function. Subsequently, specific slopes are tailored for each kernel function, applying penalties at varying rates, to enhance the model's extrapolation capabilities. Finally, this bias is seamlessly incorporated as a penalty to the post-softmax scores. We present two distinct versions of our method: a parameter-free variant that requires no new learnable parameters, which enhances length extrapolation capabilities without compromising training efficiency, and a parameterized variant capable of integrating state-of-the-art techniques. Empirical evaluations across diverse datasets have demonstrated that both variants of our method achieve state-of-the-art performance, outperforming traditional parameter-free and parameterized approaches.

★ PlainMamba: Improving Non-Hierarchical Mamba in Visual Recognition

Chenhongyi Yang, Zehui Chen, Miguel Espinosa, Linus Ericsson, Zhenyu Wang, Jiaming Liu, Elliot J. Crowley

We present PlainMamba: a simple non-hierarchical state space model (SSM) designed for general visual recognition. The recent Mamba model has shown how SSMs can be highly competitive with other architectures on sequential data and initial attempts have been made to apply it to images. In this paper, we further adapt the selective scanning process of Mamba to the visual domain, enhancing its ability to learn features from two-dimensional images by (i) a continuous 2D scanning process that improves spatial continuity by ensuring adjacency of tokens in the scanning sequence, and (ii) direction-aware updating which enables the model to discern the spatial relations of tokens by encoding directional information. Our architecture is designed to be easy to use and easy to scale, formed by stacking identical PlainMamba blocks, resulting in a model with constant width throughout all layers. The architecture is further simplified by removing the need for special tokens. We evaluate PlainMamba on a variety of visual recognition tasks including image classification, semantic segmentation, object detection, and instance segmentation. Our method achieves performance gains over previous non-hierarchical models and is competitive with hierarchical alternatives. For tasks requiring high-resolution inputs, in particular, PlainMamba requires much less computing while maintaining high performance. Code and models are available at https://github.com/ChenhongyiYang/PlainMamba

☆ Manifold-Guided Lyapunov Control with Diffusion Models

Amartya Mukherjee, Thanin Quartz, Jun Liu

This paper presents a novel approach to generating stabilizing controllers for a large class of dynamical systems using diffusion models. The core objective is to develop stabilizing control functions by identifying the closest asymptotically stable vector field relative to a predetermined manifold and adjusting the control function based on this finding. To achieve this, we employ a diffusion model trained on pairs consisting of asymptotically stable vector fields and their corresponding Lyapunov functions. Our numerical results demonstrate that this pre-trained model can achieve stabilization over previously unseen systems efficiently and rapidly, showcasing the potential of our approach in fast zero-shot control and generalizability.

comment: 14 pages

☆ How Private is DP-SGD?

Lynn Chua, Badih Ghazi, Pritish Kamath, Ravi Kumar, Pasin Manurangsi, Amer Sinha, Chiyuan Zhang

We demonstrate a substantial gap between the privacy guarantees of the Adaptive Batch Linear Queries (ABLQ) mechanism under different types of batch sampling: (i) Shuffling, and (ii) Poisson subsampling; the typical analysis of Differentially Private Stochastic Gradient Descent (DP-SGD) follows by interpreting it as a post-processing of ABLQ. While shuffling based DP-SGD is more commonly used in practical implementations, it is neither analytically nor numerically amenable to easy privacy analysis. On the other hand, Poisson subsampling based DP-SGD is challenging to scalably implement, but has a well-understood privacy analysis, with multiple open-source numerically tight privacy accountants available. This has led to a common practice of using shuffling based DP-SGD in practice, but using the privacy analysis for the corresponding Poisson subsampling version. Our result shows that there can be a substantial gap between the privacy analysis when using the two types of batch sampling, and thus advises caution in reporting privacy parameters for DP-SGD.

☆ CANOS: A Fast and Scalable Neural AC-OPF Solver Robust To N-1 Perturbations

Luis Piloto, Sofia Liguori, Sephora Madjiheurem, Miha Zgubic, Sean Lovett, Hamish Tomlinson, Sophie Elster, Chris Apps, Sims Witherspoon

Optimal Power Flow (OPF) refers to a wide range of related optimization problems with the goal of operating power systems efficiently and securely. In the simplest setting, OPF determines how much power to generate in order to minimize costs while meeting demand for power and satisfying physical and operational constraints. In even the simplest case, power grid operators use approximations of the AC-OPF problem because solving the exact problem is prohibitively slow with state-of-the-art solvers. These approximations sacrifice accuracy and operational feasibility in favor of speed. This trade-off leads to costly "uplift payments" and increased carbon emissions, especially for large power grids. In the present work, we train a deep learning system (CANOS) to predict near-optimal solutions (within 1% of the true AC-OPF cost) without compromising speed (running in as little as 33--65 ms). Importantly, CANOS scales to realistic grid sizes with promising empirical results on grids containing as many as 10,000 buses. Finally, because CANOS is a Graph Neural Network, it is robust to changes in topology. We show that CANOS is accurate across N-1 topological perturbations of a base grid typically used in security-constrained analysis. This paves the way for more efficient optimization of more complex OPF problems which alter grid connectivity such as unit commitment, topology optimization and security-constrained OPF.

☆ SGHormer: An Energy-Saving Graph Transformer Driven by Spikes

Huizhe Zhang, Jintang Li, Liang Chen, Zibin Zheng

Graph Transformers (GTs) with powerful representation learning ability make a huge success in wide range of graph tasks. However, the costs behind outstanding performances of GTs are higher energy consumption and computational overhead. The complex structure and quadratic complexity during attention calculation in vanilla transformer seriously hinder its scalability on the large-scale graph data. Though existing methods have made strides in simplifying combinations among blocks or attention-learning paradigm to improve GTs' efficiency, a series of energy-saving solutions originated from biologically plausible structures are rarely taken into consideration when constructing GT framework. To this end, we propose a new spiking-based graph transformer (SGHormer). It turns full-precision embeddings into sparse and binarized spikes to reduce memory and computational costs. The spiking graph self-attention and spiking rectify blocks in SGHormer explicitly capture global structure information and recover the expressive power of spiking embeddings, respectively. In experiments, SGHormer achieves comparable performances to other full-precision GTs with extremely low computational energy consumption. The results show that SGHomer makes a remarkable progress in the field of low-energy GTs.

comment: 9 pages, 3 figures

☆ Uncertainty-aware Distributional Offline Reinforcement Learning

Xiaocong Chen, Siyu Wang, Tong Yu, Lina Yao

Offline reinforcement learning (RL) presents distinct challenges as it relies solely on observational data. A central concern in this context is ensuring the safety of the learned policy by quantifying uncertainties associated with various actions and environmental stochasticity. Traditional approaches primarily emphasize mitigating epistemic uncertainty by learning risk-averse policies, often overlooking environmental stochasticity. In this study, we propose an uncertainty-aware distributional offline RL method to simultaneously address both epistemic uncertainty and environmental stochasticity. We propose a model-free offline RL algorithm capable of learning risk-averse policies and characterizing the entire distribution of discounted cumulative rewards, as opposed to merely maximizing the expected value of accumulated discounted returns. Our method is rigorously evaluated through comprehensive experiments in both risk-sensitive and risk-neutral benchmarks, demonstrating its superior performance.

☆ PeersimGym: An Environment for Solving the Task Offloading Problem with Reinforcement Learning

Frederico Metelo, Stevo Racković, Pedro Ákos, Cláudia Soares

Task offloading, crucial for balancing computational loads across devices in networks such as the Internet of Things, poses significant optimization challenges, including minimizing latency and energy usage under strict communication and storage constraints. While traditional optimization falls short in scalability; and heuristic approaches lack in achieving optimal outcomes, Reinforcement Learning (RL) offers a promising avenue by enabling the learning of optimal offloading strategies through iterative interactions. However, the efficacy of RL hinges on access to rich datasets and custom-tailored, realistic training environments. To address this, we introduce PeersimGym, an open-source, customizable simulation environment tailored for developing and optimizing task offloading strategies within computational networks. PeersimGym supports a wide range of network topologies and computational constraints and integrates a \textit{PettingZoo}-based interface for RL agent deployment in both solo and multi-agent setups. Furthermore, we demonstrate the utility of the environment through experiments with Deep Reinforcement Learning agents, showcasing the potential of RL-based approaches to significantly enhance offloading strategies in distributed computing settings. PeersimGym thus bridges the gap between theoretical RL models and their practical applications, paving the way for advancements in efficient task offloading methodologies.

☆ Retentive Decision Transformer with Adaptive Masking for Reinforcement Learning based Recommendation Systems

Siyu Wang, Xiaocong Chen, Lina Yao

Reinforcement Learning-based Recommender Systems (RLRS) have shown promise across a spectrum of applications, from e-commerce platforms to streaming services. Yet, they grapple with challenges, notably in crafting reward functions and harnessing large pre-existing datasets within the RL framework. Recent advancements in offline RLRS provide a solution for how to address these two challenges. However, existing methods mainly rely on the transformer architecture, which, as sequence lengths increase, can introduce challenges associated with computational resources and training costs. Additionally, the prevalent methods employ fixed-length input trajectories, restricting their capacity to capture evolving user preferences. In this study, we introduce a new offline RLRS method to deal with the above problems. We reinterpret the RLRS challenge by modeling sequential decision-making as an inference task, leveraging adaptive masking configurations. This adaptive approach selectively masks input tokens, transforming the recommendation task into an inference challenge based on varying token subsets, thereby enhancing the agent's ability to infer across diverse trajectory lengths. Furthermore, we incorporate a multi-scale segmented retention mechanism that facilitates efficient modeling of long sequences, significantly enhancing computational efficiency. Our experimental analysis, conducted on both online simulator and offline datasets, clearly demonstrates the advantages of our proposed method.

★ Data-driven Energy Consumption Modelling for Electric Micromobility using an Open Dataset

Yue Ding, Sen Yan, Maqsood Hussain Shah, Hongyuan Fang, Ji Li, Mingming Liu

The escalating challenges of traffic congestion and environmental degradation underscore the critical importance of embracing E-Mobility solutions in urban spaces. In particular, micro E-Mobility tools such as E-scooters and E-bikes, play a pivotal role in this transition, offering sustainable alternatives for urban commuters. However, the energy consumption patterns for these tools are a critical aspect that impacts their effectiveness in real-world scenarios and is essential for trip planning and boosting user confidence in using these. To this effect, recent studies have utilised physical models customised for specific mobility tools and conditions, but these models struggle with generalization and effectiveness in real-world scenarios due to a notable absence of open datasets for thorough model evaluation and verification. To fill this gap, our work presents an open dataset, collected in Dublin, Ireland, specifically designed for energy modelling research related to E-Scooters and E-Bikes. Furthermore, we provide a comprehensive analysis of energy consumption modelling based on the dataset using a set of representative machine learning algorithms and compare their performance against the contemporary mathematical models as a baseline. Our results demonstrate a notable advantage for data-driven models in comparison to the corresponding mathematical models for estimating energy consumption. Specifically, data-driven models outperform physical models in accuracy by up to 83.83% for E-Bikes and 82.16% for E-Scooters based on an in-depth analysis of the dataset under certain assumptions.

comment: 7 pages, 5 figures, 4 tables. This manuscript has been accepted by the IEEE ITEC 2024

☆ Fake or JPEG? Revealing Common Biases in Generated Image Detection Datasets

Patrick Grommelt, Louis Weiss, Franz-Josef Pfreundt, Janis Keuper

The widespread adoption of generative image models has highlighted the urgent need to detect artificial content, which is a crucial step in combating widespread manipulation and misinformation. Consequently, numerous detectors and associated datasets have emerged. However, many of these datasets inadvertently introduce undesirable biases, thereby impacting the effectiveness and evaluation of detectors. In this paper, we emphasize that many datasets for AI-generated image detection contain biases related to JPEG compression and image size. Using the GenImage dataset, we demonstrate that detectors indeed learn from these undesired factors. Furthermore, we show that removing the named biases substantially increases robustness to JPEG compression and significantly alters the cross-generator performance of evaluated detectors. Specifically, it leads to more than 11 percentage points increase in cross-generator performance for ResNet50 and Swin-T detectors on the GenImage dataset, achieving state-of-the-art results. We provide the dataset and source codes of this paper on the anonymous website: https://www.unbiased-genimage.org

☆ LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation

Ke Guo, Zhenwei Miao, Wei Jing, Weiwei Liu, Weizi Li, Dayang Hao, Jia Pan

Microscopic traffic simulation plays a crucial role in transportation engineering by providing insights into individual vehicle behavior and overall traffic flow. However, creating a realistic simulator that accurately replicates human driving behaviors in various traffic conditions presents significant challenges. Traditional simulators relying on heuristic models often fail to deliver accurate simulations due to the complexity of real-world traffic environments. Due to the covariate shift issue, existing imitation learning-based simulators often fail to generate stable long-term simulations. In this paper, we propose a novel approach called learner-aware supervised imitation learning to address the covariate shift problem in multi-agent imitation learning. By leveraging a variational autoencoder simultaneously modeling the expert and learner state distribution, our approach augments expert states such that the augmented state is aware of learner state distribution. Our method, applied to urban traffic simulation, demonstrates significant improvements over existing state-of-the-art baselines in both short-term microscopic and long-term macroscopic realism when evaluated on the real-world dataset pNEUMA.

comment: accepted by cvpr 2024. arXiv admin note: text overlap with arXiv:2306.06401

☆ On the Benefits of Over-parameterization for Out-of-Distribution Generalization

Yifan Hao, Yong Lin, Difan Zou, Tong Zhang

In recent years, machine learning models have achieved success based on the independently and identically distributed assumption. However, this assumption can be easily violated in real-world applications, leading to the Out-of-Distribution (OOD) problem. Understanding how modern over-parameterized DNNs behave under non-trivial natural distributional shifts is essential, as current theoretical understanding is insufficient. Existing theoretical works often provide meaningless results for over-parameterized models in OOD scenarios or even contradict empirical findings. To this end, we are investigating the performance of the over-parameterized model in terms of OOD generalization under the general benign overfitting conditions. Our analysis focuses on a random feature model and examines non-trivial natural distributional shifts, where the benign overfitting estimators demonstrate a constant excess OOD loss, despite achieving zero excess in-distribution (ID) loss. We demonstrate that in this scenario, further increasing the model's parameterization can significantly reduce the OOD loss. Intuitively, the variance term of ID loss remains low due to orthogonality of long-tail features, meaning overfitting noise during training generally doesn't raise testing loss. However, in OOD cases, distributional shift increases the variance term. Thankfully, the inherent shift is unrelated to individual x, maintaining the orthogonality of long-tail features. Expanding the hidden dimension can additionally improve this orthogonality by mapping the features into higher-dimensional spaces, thereby reducing the variance term. We further show that model ensembles also improve OOD loss, akin to increasing model capacity. These insights explain the empirical phenomenon of enhanced OOD generalization through model ensembles, supported by consistent simulations with theoretical results.

☆ Forest-ORE: Mining Optimal Rule Ensemble to interpret Random Forest models

Haddouchi Maissae, Berrado Abdelaziz

Random Forest (RF) is well-known as an efficient ensemble learning method in terms of predictive performance. It is also considered a Black Box because of its hundreds of deep decision trees. This lack of interpretability can be a real drawback for acceptance of RF models in several real-world applications, especially those affecting one's lives, such as in healthcare, security, and law. In this work, we present Forest-ORE, a method that makes RF interpretable via an optimized rule ensemble (ORE) for local and global interpretation. Unlike other rule-based approaches aiming at interpreting the RF model, this method simultaneously considers several parameters that influence the choice of an interpretable rule ensemble. Existing methods often prioritize predictive performance over interpretability coverage and do not provide information about existing overlaps or interactions between rules. Forest-ORE uses a mixed-integer optimization program to build an ORE that considers the trade-off between predictive performance, interpretability coverage, and model size (size of the rule ensemble, rule lengths, and rule overlaps). In addition to providing an ORE competitive in predictive performance with RF, this method enriches the ORE through other rules that afford complementary information. It also enables monitoring of the rule selection process and delivers various metrics that can be used to generate a graphical representation of the final model. This framework is illustrated through an example, and its robustness is assessed through 36 benchmark datasets. A comparative analysis of well-known methods shows that Forest-ORE provides an excellent trade-off between predictive performance, interpretability coverage, and model size.

comment: 48 pages, 11 figures

☆ Dual Memory Networks: A Versatile Adaptation Approach for Vision-Language Models CVPR2024

Yabin Zhang, Wenjie Zhu, Hui Tang, Zhiyuan Ma, Kaiyang Zhou, Lei Zhang

With the emergence of pre-trained vision-language models like CLIP, how to adapt them to various downstream classification tasks has garnered significant attention in recent research. The adaptation strategies can be typically categorized into three paradigms: zero-shot adaptation, few-shot adaptation, and the recently-proposed training-free few-shot adaptation. Most existing approaches are tailored for a specific setting and can only cater to one or two of these paradigms. In this paper, we introduce a versatile adaptation approach that can effectively work under all three settings. Specifically, we propose the dual memory networks that comprise dynamic and static memory components. The static memory caches training data knowledge, enabling training-free few-shot adaptation, while the dynamic memory preserves historical test features online during the testing process, allowing for the exploration of additional data insights beyond the training set. This novel capability enhances model performance in the few-shot setting and enables model usability in the absence of training data. The two memory networks employ the same flexible memory interactive strategy, which can operate in a training-free mode and can be further enhanced by incorporating learnable projection layers. Our approach is tested across 11 datasets under the three task settings. Remarkably, in the zero-shot scenario, it outperforms existing methods by over 3\% and even shows superior results against methods utilizing external training data. Additionally, our method exhibits robust performance against natural distribution shifts. Codes are available at \url{https://github.com/YBZh/DMN}.

comment: CVPR2024; Codes are available at \url{https://github.com/YBZh/DMN}

☆ Towards a Zero-Data, Controllable, Adaptive Dialog System

Dirk Väth, Lindsey Vanderlyn, Ngoc Thang Vu

Conversational Tree Search (V\"ath et al., 2023) is a recent approach to controllable dialog systems, where domain experts shape the behavior of a Reinforcement Learning agent through a dialog tree. The agent learns to efficiently navigate this tree, while adapting to information needs, e.g., domain familiarity, of different users. However, the need for additional training data hinders deployment in new domains. To address this, we explore approaches to generate this data directly from dialog trees. We improve the original approach, and show that agents trained on synthetic data can achieve comparable dialog success to models trained on human data, both when using a commercial Large Language Model for generation, or when using a smaller open-source model, running on a single GPU. We further demonstrate the scalability of our approach by collecting and testing on two new datasets: ONBOARD, a new domain helping foreign residents moving to a new city, and the medical domain DIAGNOSE, a subset of Wikipedia articles related to scalp and head symptoms. Finally, we perform human testing, where no statistically significant differences were found in either objective or subjective measures between models trained on human and generated data.

☆ Enhancing Privacy in Federated Learning through Local Training

Nicola Bastianello, Changxin Liu, Karl H. Johansson

In this paper we propose the federated private local training algorithm (Fed-PLT) for federated learning, to overcome the challenges of (i) expensive communications and (ii) privacy preservation. We address (i) by allowing for both partial participation and local training, which significantly reduce the number of communication rounds between the central coordinator and computing agents. The algorithm matches the state of the art in the sense that the use of local training demonstrably does not impact accuracy. Additionally, agents have the flexibility to choose from various local training solvers, such as (stochastic) gradient descent and accelerated gradient descent. Further, we investigate how employing local training can enhance privacy, addressing point (ii). In particular, we derive differential privacy bounds and highlight their dependence on the number of local training epochs. We assess the effectiveness of the proposed algorithm by comparing it to alternative techniques, considering both theoretical analysis and numerical results from a classification task.

☆ A Survey on Deep Learning and State-of-the-arts Applications

Mohd Halim Mohd Noor, Ayokunle Olalekan Ige

Deep learning, a branch of artificial intelligence, is a computational model that uses multiple layers of interconnected units (neurons) to learn intricate patterns and representations directly from raw input data. Empowered by this learning capability, it has become a powerful tool for solving complex problems and is the core driver of many groundbreaking technologies and innovations. Building a deep learning model is a challenging task due to the algorithm`s complexity and the dynamic nature of real-world problems. Several studies have reviewed deep learning concepts and applications. However, the studies mostly focused on the types of deep learning models and convolutional neural network architectures, offering limited coverage of the state-of-the-art of deep learning models and their applications in solving complex problems across different domains. Therefore, motivated by the limitations, this study aims to comprehensively review the state-of-the-art deep learning models in computer vision, natural language processing, time series analysis and pervasive computing. We highlight the key features of the models and their effectiveness in solving the problems within each domain. Furthermore, this study presents the fundamentals of deep learning, various deep learning model types and prominent convolutional neural network architectures. Finally, challenges and future directions in deep learning research are discussed to offer a broader perspective for future researchers.

☆ DeepMIF: Deep Monotonic Implicit Fields for Large-Scale LiDAR 3D Mapping

Kutay Yılmaz, Matthias Nießner, Anastasiia Kornilova, Alexey Artemov

Recently, significant progress has been achieved in sensing real large-scale outdoor 3D environments, particularly by using modern acquisition equipment such as LiDAR sensors. Unfortunately, they are fundamentally limited in their ability to produce dense, complete 3D scenes. To address this issue, recent learning-based methods integrate neural implicit representations and optimizable feature grids to approximate surfaces of 3D scenes. However, naively fitting samples along raw LiDAR rays leads to noisy 3D mapping results due to the nature of sparse, conflicting LiDAR measurements. Instead, in this work we depart from fitting LiDAR data exactly, instead letting the network optimize a non-metric monotonic implicit field defined in 3D space. To fit our field, we design a learning system integrating a monotonicity loss that enables optimizing neural monotonic fields and leverages recent progress in large-scale 3D mapping. Our algorithm achieves high-quality dense 3D mapping performance as captured by multiple quantitative and perceptual measures and visual results obtained for Mai City, Newer College, and KITTI benchmarks. The code of our approach will be made publicly available.

comment: 8 pages, 6 figures

☆ VDSC: Enhancing Exploration Timing with Value Discrepancy and State Counts

Marius Captari, Remo Sasso, Matthia Sabatelli

Despite the considerable attention given to the questions of \textit{how much} and \textit{how to} explore in deep reinforcement learning, the investigation into \textit{when} to explore remains relatively less researched. While more sophisticated exploration strategies can excel in specific, often sparse reward environments, existing simpler approaches, such as $\epsilon$-greedy, persist in outperforming them across a broader spectrum of domains. The appeal of these simpler strategies lies in their ease of implementation and generality across a wide range of domains. The downside is that these methods are essentially a blind switching mechanism, which completely disregards the agent's internal state. In this paper, we propose to leverage the agent's internal state to decide \textit{when} to explore, addressing the shortcomings of blind switching mechanisms. We present Value Discrepancy and State Counts through homeostasis (VDSC), a novel approach for efficient exploration timing. Experimental results on the Atari suite demonstrate the superiority of our strategy over traditional methods such as $\epsilon$-greedy and Boltzmann, as well as more sophisticated techniques like Noisy Nets.

☆ BVR Gym: A Reinforcement Learning Environment for Beyond-Visual-Range Air Combat

Edvards Scukins, Markus Klein, Lars Kroon, Petter Ögren

Creating new air combat tactics and discovering novel maneuvers can require numerous hours of expert pilots' time. Additionally, for each different combat scenario, the same strategies may not work since small changes in equipment performance may drastically change the air combat outcome. For this reason, we created a reinforcement learning environment to help investigate potential air combat tactics in the field of beyond-visual-range (BVR) air combat: the BVR Gym. This type of air combat is important since long-range missiles are often the first weapon to be used in aerial combat. Some existing environments provide high-fidelity simulations but are either not open source or are not adapted to the BVR air combat domain. Other environments are open source but use less accurate simulation models. Our work provides a high-fidelity environment based on the open-source flight dynamics simulator JSBSim and is adapted to the BVR air combat domain. This article describes the building blocks of the environment and some use cases.

comment: 8 pages, 8 figures

☆ Boosting Adversarial Training via Fisher-Rao Norm-based Regularization CVPR2024

Xiangyu Yin, Wenjie Ruan

Adversarial training is extensively utilized to improve the adversarial robustness of deep neural networks. Yet, mitigating the degradation of standard generalization performance in adversarial-trained models remains an open problem. This paper attempts to resolve this issue through the lens of model complexity. First, We leverage the Fisher-Rao norm, a geometrically invariant metric for model complexity, to establish the non-trivial bounds of the Cross-Entropy Loss-based Rademacher complexity for a ReLU-activated Multi-Layer Perceptron. Then we generalize a complexity-related variable, which is sensitive to the changes in model width and the trade-off factors in adversarial training. Moreover, intensive empirical evidence validates that this variable highly correlates with the generalization gap of Cross-Entropy loss between adversarial-trained and standard-trained models, especially during the initial and final phases of the training process. Building upon this observation, we propose a novel regularization framework, called Logit-Oriented Adversarial Training (LOAT), which can mitigate the trade-off between robustness and accuracy while imposing only a negligible increase in computational overhead. Our extensive experiments demonstrate that the proposed regularization strategy can boost the performance of the prevalent adversarial training algorithms, including PGD-AT, TRADES, TRADES (LSE), MART, and DM-AT, across various network architectures. Our code will be available at https://github.com/TrustAI/LOAT.

comment: This paper has been accepted to CVPR2024

☆ Prediction-sharing During Training and Inference

Yotam Gafni, Ronen Gradwohl, Moshe Tennenholtz

Two firms are engaged in a competitive prediction task. Each firm has two sources of data -- labeled historical data and unlabeled inference-time data -- and uses the former to derive a prediction model, and the latter to make predictions on new instances. We study data-sharing contracts between the firms. The novelty of our study is to introduce and highlight the differences between contracts that share prediction models only, contracts to share inference-time predictions only, and contracts to share both. Our analysis proceeds on three levels. First, we develop a general Bayesian framework that facilitates our study. Second, we narrow our focus to two natural settings within this framework: (i) a setting in which the accuracy of each firm's prediction model is common knowledge, but the correlation between the respective models is unknown; and (ii) a setting in which two hypotheses exist regarding the optimal predictor, and one of the firms has a structural advantage in deducing it. Within these two settings we study optimal contract choice. More specifically, we find the individually rational and Pareto-optimal contracts for some notable cases, and describe specific settings where each of the different sharing contracts emerge as optimal. Finally, in the third level of our analysis we demonstrate the applicability of our concepts in a synthetic simulation using real loan data.

☆ EL-MLFFs: Ensemble Learning of Machine Leaning Force Fields

Bangchen Yin, Yue Yin, Yuda W. Tang, Hai Xiao

Machine learning force fields (MLFFs) have emerged as a promising approach to bridge the accuracy of quantum mechanical methods and the efficiency of classical force fields. However, the abundance of MLFF models and the challenge of accurately predicting atomic forces pose significant obstacles in their practical application. In this paper, we propose a novel ensemble learning framework, EL-MLFFs, which leverages the stacking method to integrate predictions from diverse MLFFs and enhance force prediction accuracy. By constructing a graph representation of molecular structures and employing a graph neural network (GNN) as the meta-model, EL-MLFFs effectively captures atomic interactions and refines force predictions. We evaluate our approach on two distinct datasets: methane molecules and methanol adsorbed on a Cu(100) surface. The results demonstrate that EL-MLFFs significantly improves force prediction accuracy compared to individual MLFFs, with the ensemble of all eight models yielding the best performance. Moreover, our ablation study highlights the crucial roles of the residual network and graph attention layers in the model's architecture. The EL-MLFFs framework offers a promising solution to the challenges of model selection and force prediction accuracy in MLFFs, paving the way for more reliable and efficient molecular simulations.

comment: 12 pages, 3 figures

☆ DS-AL: A Dual-Stream Analytic Learning for Exemplar-Free Class-Incremental Learning AAAI 2024

Huiping Zhuang, Run He, Kai Tong, Ziqian Zeng, Cen Chen, Zhiping Lin

Class-incremental learning (CIL) under an exemplar-free constraint has presented a significant challenge. Existing methods adhering to this constraint are prone to catastrophic forgetting, far more so than replay-based techniques that retain access to past samples. In this paper, to solve the exemplar-free CIL problem, we propose a Dual-Stream Analytic Learning (DS-AL) approach. The DS-AL contains a main stream offering an analytical (i.e., closed-form) linear solution, and a compensation stream improving the inherent under-fitting limitation due to adopting linear mapping. The main stream redefines the CIL problem into a Concatenated Recursive Least Squares (C-RLS) task, allowing an equivalence between the CIL and its joint-learning counterpart. The compensation stream is governed by a Dual-Activation Compensation (DAC) module. This module re-activates the embedding with a different activation function from the main stream one, and seeks fitting compensation by projecting the embedding to the null space of the main stream's linear mapping. Empirical results demonstrate that the DS-AL, despite being an exemplar-free technique, delivers performance comparable with or better than that of replay-based methods across various datasets, including CIFAR-100, ImageNet-100 and ImageNet-Full. Additionally, the C-RLS' equivalent property allows the DS-AL to execute CIL in a phase-invariant manner. This is evidenced by a never-before-seen 500-phase CIL ImageNet task, which performs on a level identical to a 5-phase one. Our codes are available at https://github.com/ZHUANGHP/Analytic-continual-learning.

comment: Accepted in AAAI 2024

☆ Variational Graph Auto-Encoder Based Inductive Learning Method for Semi-Supervised Classification

Hanxuan Yang, Zhaoxin Yu, Qingchao Kong, Wei Liu, Wenji Mao

Graph representation learning is a fundamental research issue in various domains of applications, of which the inductive learning problem is particularly challenging as it requires models to generalize to unseen graph structures during inference. In recent years, graph neural networks (GNNs) have emerged as powerful graph models for inductive learning tasks such as node classification, whereas they typically heavily rely on the annotated nodes under a fully supervised training setting. Compared with the GNN-based methods, variational graph auto-encoders (VGAEs) are known to be more generalizable to capture the internal structural information of graphs independent of node labels and have achieved prominent performance on multiple unsupervised learning tasks. However, so far there is still a lack of work focusing on leveraging the VGAE framework for inductive learning, due to the difficulties in training the model in a supervised manner and avoiding over-fitting the proximity information of graphs. To solve these problems and improve the model performance of VGAEs for inductive graph representation learning, in this work, we propose the Self-Label Augmented VGAE model. To leverage the label information for training, our model takes node labels as one-hot encoded inputs and then performs label reconstruction in model training. To overcome the scarcity problem of node labels for semi-supervised settings, we further propose the Self-Label Augmentation Method (SLAM), which uses pseudo labels generated by our model with a node-wise masking approach to enhance the label information. Experiments on benchmark inductive learning graph datasets verify that our proposed model archives promising results on node classification with particular superiority under semi-supervised learning settings.

☆ Capacity Provisioning Motivated Online Non-Convex Optimization Problem with Memory and Switching Cost

Rahul Vaze, Jayakrishnan Nair

An online non-convex optimization problem is considered where the goal is to minimize the flow time (total delay) of a set of jobs by modulating the number of active servers, but with a switching cost associated with changing the number of active servers over time. Each job can be processed by at most one fixed speed server at any time. Compared to the usual online convex optimization (OCO) problem with switching cost, the objective function considered is non-convex and more importantly, at each time, it depends on all past decisions and not just the present one. Both worst-case and stochastic inputs are considered; for both cases, competitive algorithms are derived.

☆ Natural Language Requirements Testability Measurement Based on Requirement Smells

Morteza Zakeri-Nasrabadi, Saeed Parsa

Requirements form the basis for defining software systems' obligations and tasks. Testable requirements help prevent failures, reduce maintenance costs, and make it easier to perform acceptance tests. However, despite the importance of measuring and quantifying requirements testability, no automatic approach for measuring requirements testability has been proposed based on the requirements smells, which are at odds with the requirements testability. This paper presents a mathematical model to evaluate and rank the natural language requirements testability based on an extensive set of nine requirements smells, detected automatically, and acceptance test efforts determined by requirement length and its application domain. Most of the smells stem from uncountable adjectives, context-sensitive, and ambiguous words. A comprehensive dictionary is required to detect such words. We offer a neural word-embedding technique to generate such a dictionary automatically. Using the dictionary, we could automatically detect Polysemy smell (domain-specific ambiguity) for the first time in 10 application domains. Our empirical study on nearly 1000 software requirements from six well-known industrial and academic projects demonstrates that the proposed smell detection approach outperforms Smella, a state-of-the-art tool, in detecting requirements smells. The precision and recall of smell detection are improved with an average of 0.03 and 0.33, respectively, compared to the state-of-the-art. The proposed requirement testability model measures the testability of 985 requirements with a mean absolute error of 0.12 and a mean squared error of 0.03, demonstrating the model's potential for practical use.

comment: 45 pages, 16 figures, and 13 tables; submitted as a journal paper

☆ A Unified Kernel for Neural Network Learning

Shao-Qun Zhang, Zong-Yi Chen, Yong-Ming Tian, Xun Lu

Past decades have witnessed a great interest in the distinction and connection between neural network learning and kernel learning. Recent advancements have made theoretical progress in connecting infinite-wide neural networks and Gaussian processes. Two predominant approaches have emerged: the Neural Network Gaussian Process (NNGP) and the Neural Tangent Kernel (NTK). The former, rooted in Bayesian inference, represents a zero-order kernel, while the latter, grounded in the tangent space of gradient descents, is a first-order kernel. In this paper, we present the Unified Neural Kernel (UNK), which characterizes the learning dynamics of neural networks with gradient descents and parameter initialization. The proposed UNK kernel maintains the limiting properties of both NNGP and NTK, exhibiting behaviors akin to NTK with a finite learning step and converging to NNGP as the learning step approaches infinity. Besides, we also theoretically characterize the uniform tightness and learning convergence of the UNK kernel, providing comprehensive insights into this unified kernel. Experimental results underscore the effectiveness of our proposed method.

☆ Chain of Compression: A Systematic Approach to Combinationally Compress Convolutional Neural Networks

Yingtao Shen, Minqing Sun, Jie Zhao, An Zou

Convolutional neural networks (CNNs) have achieved significant popularity, but their computational and memory intensity poses challenges for resource-constrained computing systems, particularly with the prerequisite of real-time performance. To release this burden, model compression has become an important research focus. Many approaches like quantization, pruning, early exit, and knowledge distillation have demonstrated the effect of reducing redundancy in neural networks. Upon closer examination, it becomes apparent that each approach capitalizes on its unique features to compress the neural network, and they can also exhibit complementary behavior when combined. To explore the interactions and reap the benefits from the complementary features, we propose the Chain of Compression, which works on the combinational sequence to apply these common techniques to compress the neural network. Validated on the image-based regression and classification networks across different data sets, our proposed Chain of Compression can significantly compress the computation cost by 100-1000 times with ignorable accuracy loss compared with the baseline model.

comment: 10 pages, 15 figures

☆ Incorporating Exponential Smoothing into MLP: A Simple but Effective Sequence Model

Jiqun Chu, Zuoquan Lin

Modeling long-range dependencies in sequential data is a crucial step in sequence learning. A recently developed model, the Structured State Space (S4), demonstrated significant effectiveness in modeling long-range sequences. However, It is unclear whether the success of S4 can be attributed to its intricate parameterization and HiPPO initialization or simply due to State Space Models (SSMs). To further investigate the potential of the deep SSMs, we start with exponential smoothing (ETS), a simple SSM, and propose a stacked architecture by directly incorporating it into an element-wise MLP. We augment simple ETS with additional parameters and complex field to reduce the inductive bias. Despite increasing less than 1\% of parameters of element-wise MLP, our models achieve comparable results to S4 on the LRA benchmark.

comment: 12 pages, 5 tables, 3 figures

☆ Particle identification with machine learning from incomplete data in the ALICE experiment

Maja Karwowska, Łukasz Graczykowski, Kamil Deja, Miłosz Kasak, Małgorzata Janik

The ALICE experiment at the LHC measures properties of the strongly interacting matter formed in ultrarelativistic heavy-ion collisions. Such studies require accurate particle identification (PID). ALICE provides PID information via several detectors for particles with momentum from about 100 MeV/c up to 20 GeV/c. Traditionally, particles are selected with rectangular cuts. Acmuch better performance can be achieved with machine learning (ML) methods. Our solution uses multiple neural networks (NN) serving as binary classifiers. Moreover, we extended our particle classifier with Feature Set Embedding and attention in order to train on data with incomplete samples. We also present the integration of the ML project with the ALICE analysis software, and we discuss domain adaptation, the ML technique needed to transfer the knowledge between simulated and real experimental data.

comment: Proceedings of 3rd Artificial Intelligence for the Electron Ion Collider workshop -- AI4EIC2023, 28.11-1.12.2023. Prepared for submission to JINST

☆ Robust and Scalable Model Editing for Large Language Models LREC

Yingfa Chen, Zhengyan Zhang, Xu Han, Chaojun Xiao, Zhiyuan Liu, Chen Chen, Kuai Li, Tao Yang, Maosong Sun

Large language models (LLMs) can make predictions using parametric knowledge--knowledge encoded in the model weights--or contextual knowledge--knowledge presented in the context. In many scenarios, a desirable behavior is that LLMs give precedence to contextual knowledge when it conflicts with the parametric knowledge, and fall back to using their parametric knowledge when the context is irrelevant. This enables updating and correcting the model's knowledge by in-context editing instead of retraining. Previous works have shown that LLMs are inclined to ignore contextual knowledge and fail to reliably fall back to parametric knowledge when presented with irrelevant context. In this work, we discover that, with proper prompting methods, instruction-finetuned LLMs can be highly controllable by contextual knowledge and robust to irrelevant context. Utilizing this feature, we propose EREN (Edit models by REading Notes) to improve the scalability and robustness of LLM editing. To better evaluate the robustness of model editors, we collect a new dataset, that contains irrelevant questions that are more challenging than the ones in existing datasets. Empirical results show that our method outperforms current state-of-the-art methods by a large margin. Unlike existing techniques, it can integrate knowledge from multiple edits, and correctly respond to syntactically similar but semantically unrelated inputs (and vice versa). The source code can be found at https://github.com/thunlp/EREN.

comment: LREC-COLING 2024 paper, 16 pages, 4 figures

☆ Masked Multi-Domain Network: Multi-Type and Multi-Scenario Conversion Rate Prediction with a Single Model CIKM 2023

Wentao Ouyang, Xiuwu Zhang, Chaofeng Guo, Shukui Ren, Yupei Sui, Kun Zhang, Jinmei Luo, Yunfeng Chen, Dongbo Xu, Xiangzheng Liu, Yanlong Du

In real-world advertising systems, conversions have different types in nature and ads can be shown in different display scenarios, both of which highly impact the actual conversion rate (CVR). This results in the multi-type and multi-scenario CVR prediction problem. A desired model for this problem should satisfy the following requirements: 1) Accuracy: the model should achieve fine-grained accuracy with respect to any conversion type in any display scenario. 2) Scalability: the model parameter size should be affordable. 3) Convenience: the model should not require a large amount of effort in data partitioning, subset processing and separate storage. Existing approaches cannot simultaneously satisfy these requirements. For example, building a separate model for each (conversion type, display scenario) pair is neither scalable nor convenient. Building a unified model trained on all the data with conversion type and display scenario included as two features is not accurate enough. In this paper, we propose the Masked Multi-domain Network (MMN) to solve this problem. To achieve the accuracy requirement, we model domain-specific parameters and propose a dynamically weighted loss to account for the loss scale imbalance issue within each mini-batch. To achieve the scalability requirement, we propose a parameter sharing and composition strategy to reduce model parameters from a product space to a sum space. To achieve the convenience requirement, we propose an auto-masking strategy which can take mixed data from all the domains as input. It avoids the overhead caused by data partitioning, individual processing and separate storage. Both offline and online experimental results validate the superiority of MMN for multi-type and multi-scenario CVR prediction. MMN is now the serving model for real-time CVR prediction in UC Toutiao.

comment: CIKM 2023 (larger figures)

☆ On permutation-invariant neural networks

Masanari Kimura, Ryotaro Shimizu, Yuki Hirakawa, Ryosuke Goto, Yuki Saito

Conventional machine learning algorithms have traditionally been designed under the assumption that input data follows a vector-based format, with an emphasis on vector-centric paradigms. However, as the demand for tasks involving set-based inputs has grown, there has been a paradigm shift in the research community towards addressing these challenges. In recent years, the emergence of neural network architectures such as Deep Sets and Transformers has presented a significant advancement in the treatment of set-based data. These architectures are specifically engineered to naturally accommodate sets as input, enabling more effective representation and processing of set structures. Consequently, there has been a surge of research endeavors dedicated to exploring and harnessing the capabilities of these architectures for various tasks involving the approximation of set functions. This comprehensive survey aims to provide an overview of the diverse problem settings and ongoing research efforts pertaining to neural networks that approximate set functions. By delving into the intricacies of these approaches and elucidating the associated challenges, the survey aims to equip readers with a comprehensive understanding of the field. Through this comprehensive perspective, we hope that researchers can gain valuable insights into the potential applications, inherent limitations, and future directions of set-based neural networks. Indeed, from this survey we gain two insights: i) Deep Sets and its variants can be generalized by differences in the aggregation function, and ii) the behavior of Deep Sets is sensitive to the choice of the aggregation function. From these observations, we show that Deep Sets, one of the well-known permutation-invariant neural networks, can be generalized in the sense of a quasi-arithmetic mean.

☆ Transcribing Bengali Text with Regional Dialects to IPA using District Guided Tokens

S M Jishanul Islam, Sadia Ahmmed, Sahid Hossain Mustakim

Accurate transcription of Bengali text to the International Phonetic Alphabet (IPA) is a challenging task due to the complex phonology of the language and context-dependent sound changes. This challenge is even more for regional Bengali dialects due to unavailability of standardized spelling conventions for these dialects, presence of local and foreign words popular in those regions and phonological diversity across different regions. This paper presents an approach to this sequence-to-sequence problem by introducing the District Guided Tokens (DGT) technique on a new dataset spanning six districts of Bangladesh. The key idea is to provide the model with explicit information about the regional dialect or "district" of the input text before generating the IPA transcription. This is achieved by prepending a district token to the input sequence, effectively guiding the model to understand the unique phonetic patterns associated with each district. The DGT technique is applied to fine-tune several transformer-based models, on this new dataset. Experimental results demonstrate the effectiveness of DGT, with the ByT5 model achieving superior performance over word-based models like mT5, BanglaT5, and umT5. This is attributed to ByT5's ability to handle a high percentage of out-of-vocabulary words in the test set. The proposed approach highlights the importance of incorporating regional dialect information into ubiquitous natural language processing systems for languages with diverse phonological variations. The following work was a result of the "Bhashamul" challenge, which is dedicated to solving the problem of Bengali text with regional dialects to IPA transcription https://www.kaggle.com/competitions/regipa/. The training and inference notebooks are available through the competition link.

comment: This work became the champion of the Bhashamul challenge

☆ Generalization Error Analysis for Sparse Mixture-of-Experts: A Preliminary Study

Jinze Zhao, Peihao Wang, Zhangyang Wang

Mixture-of-Experts (MoE) represents an ensemble methodology that amalgamates predictions from several specialized sub-models (referred to as experts). This fusion is accomplished through a router mechanism, dynamically assigning weights to each expert's contribution based on the input data. Conventional MoE mechanisms select all available experts, incurring substantial computational costs. In contrast, Sparse Mixture-of-Experts (Sparse MoE) selectively engages only a limited number, or even just one expert, significantly reducing computation overhead while empirically preserving, and sometimes even enhancing, performance. Despite its wide-ranging applications and these advantageous characteristics, MoE's theoretical underpinnings have remained elusive. In this paper, we embark on an exploration of Sparse MoE's generalization error concerning various critical factors. Specifically, we investigate the impact of the number of data samples, the total number of experts, the sparsity in expert selection, the complexity of the routing mechanism, and the complexity of individual experts. Our analysis sheds light on \textit{how \textbf{sparsity} contributes to the MoE's generalization}, offering insights from the perspective of classical learning theory.

☆ Application-Driven Innovation in Machine Learning

David Rolnick, Alan Aspuru-Guzik, Sara Beery, Bistra Dilkina, Priya L. Donti, Marzyeh Ghassemi, Hannah Kerner, Claire Monteleoni, Esther Rolf, Milind Tambe, Adam White

As applications of machine learning proliferate, innovative algorithms inspired by specific real-world challenges have become increasingly important. Such work offers the potential for significant impact not merely in domains of application but also in machine learning itself. In this paper, we describe the paradigm of application-driven research in machine learning, contrasting it with the more standard paradigm of methods-driven research. We illustrate the benefits of application-driven machine learning and how this approach can productively synergize with methods-driven work. Despite these benefits, we find that reviewing, hiring, and teaching practices in machine learning often hold back application-driven innovation. We outline how these processes may be improved.

comment: 12 pages, 3 figures

☆ Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance

Donghoon Ahn, Hyoungwon Cho, Jaewon Min, Wooseok Jang, Jungwoo Kim, SeonHwa Kim, Hyun Hee Park, Kyong Hwan Jin, Seungryong Kim

Recent studies have demonstrated that diffusion models are capable of generating high-quality samples, but their quality heavily depends on sampling guidance techniques, such as classifier guidance (CG) and classifier-free guidance (CFG). These techniques are often not applicable in unconditional generation or in various downstream tasks such as image restoration. In this paper, we propose a novel sampling guidance, called Perturbed-Attention Guidance (PAG), which improves diffusion sample quality across both unconditional and conditional settings, achieving this without requiring additional training or the integration of external modules. PAG is designed to progressively enhance the structure of samples throughout the denoising process. It involves generating intermediate samples with degraded structure by substituting selected self-attention maps in diffusion U-Net with an identity matrix, by considering the self-attention mechanisms' ability to capture structural information, and guiding the denoising process away from these degraded samples. In both ADM and Stable Diffusion, PAG surprisingly improves sample quality in conditional and even unconditional scenarios. Moreover, PAG significantly improves the baseline performance in various downstream tasks where existing guidances such as CG or CFG cannot be fully utilized, including ControlNet with empty prompts and image restoration such as inpainting and deblurring.

comment: Project page is available at https://ku-cvlab.github.io/Perturbed-Attention-Guidance

☆ AIDE: An Automatic Data Engine for Object Detection in Autonomous Driving CVPR-2024

Mingfu Liang, Jong-Chyi Su, Samuel Schulter, Sparsh Garg, Shiyu Zhao, Ying Wu, Manmohan Chandraker

Autonomous vehicle (AV) systems rely on robust perception models as a cornerstone of safety assurance. However, objects encountered on the road exhibit a long-tailed distribution, with rare or unseen categories posing challenges to a deployed perception model. This necessitates an expensive process of continuously curating and annotating data with significant human effort. We propose to leverage recent advances in vision-language and large language models to design an Automatic Data Engine (AIDE) that automatically identifies issues, efficiently curates data, improves the model through auto-labeling, and verifies the model through generation of diverse scenarios. This process operates iteratively, allowing for continuous self-improvement of the model. We further establish a benchmark for open-world detection on AV datasets to comprehensively evaluate various learning paradigms, demonstrating our method's superior performance at a reduced cost.

comment: Accepted by CVPR-2024

☆ A Moreau Envelope Approach for LQR Meta-Policy Estimation

Ashwin Aravind, Mohammad Taha Toghani, César A. Uribe

We study the problem of policy estimation for the Linear Quadratic Regulator (LQR) in discrete-time linear time-invariant uncertain dynamical systems. We propose a Moreau Envelope-based surrogate LQR cost, built from a finite set of realizations of the uncertain system, to define a meta-policy efficiently adjustable to new realizations. Moreover, we design an algorithm to find an approximate first-order stationary point of the meta-LQR cost function. Numerical results show that the proposed approach outperforms naive averaging of controllers on new realizations of the linear system. We also provide empirical evidence that our method has better sample complexity than Model-Agnostic Meta-Learning (MAML) approaches.

comment: 8 pages

☆ Multi-Objective Trajectory Planning with Dual-Encoder

Beibei Zhang, Tian Xiang, Chentao Mao, Yuhua Zheng, Shuai Li, Haoyi Niu, Xiangming Xi, Wenyuan Bai, Feng Gao

Time-jerk optimal trajectory planning is crucial in advancing robotic arms' performance in dynamic tasks. Traditional methods rely on solving complex nonlinear programming problems, bringing significant delays in generating optimized trajectories. In this paper, we propose a two-stage approach to accelerate time-jerk optimal trajectory planning. Firstly, we introduce a dual-encoder based transformer model to establish a good preliminary trajectory. This trajectory is subsequently refined through sequential quadratic programming to improve its optimality and robustness. Our approach outperforms the state-of-the-art by up to 79.72\% in reducing trajectory planning time. Compared with existing methods, our method shrinks the optimality gap with the objective function value decreasing by up to 29.9\%.

comment: 6 pages, 7 figures, conference

☆ Learn from Heterophily: Heterophilous Information-enhanced Graph Neural Network

Yilun Zheng, Jiahao Xu, Lihui Chen

Under circumstances of heterophily, where nodes with different labels tend to be connected based on semantic meanings, Graph Neural Networks (GNNs) often exhibit suboptimal performance. Current studies on graph heterophily mainly focus on aggregation calibration or neighbor extension and address the heterophily issue by utilizing node features or structural information to improve GNN representations. In this paper, we propose and demonstrate that the valuable semantic information inherent in heterophily can be utilized effectively in graph learning by investigating the distribution of neighbors for each individual node within the graph. The theoretical analysis is carried out to demonstrate the efficacy of the idea in enhancing graph learning. Based on this analysis, we propose HiGNN, an innovative approach that constructs an additional new graph structure, that integrates heterophilous information by leveraging node distribution to enhance connectivity between nodes that share similar semantic characteristics. We conduct empirical assessments on node classification tasks using both homophilous and heterophilous benchmark datasets and compare HiGNN to popular GNN baselines and SoTA methods, confirming the effectiveness in improving graph representations. In addition, by incorporating heterophilous information, we demonstrate a notable enhancement in existing GNN-based approaches, and the homophily degree across real-world datasets, thus affirming the efficacy of our approach.

☆ The Pursuit of Fairness in Artificial Intelligence Models: A Survey

Tahsin Alamgir Kheya, Mohamed Reda Bouadjenek, Sunil Aryal

Artificial Intelligence (AI) models are now being utilized in all facets of our lives such as healthcare, education and employment. Since they are used in numerous sensitive environments and make decisions that can be life altering, potential biased outcomes are a pressing matter. Developers should ensure that such models don't manifest any unexpected discriminatory practices like partiality for certain genders, ethnicities or disabled people. With the ubiquitous dissemination of AI systems, researchers and practitioners are becoming more aware of unfair models and are bound to mitigate bias in them. Significant research has been conducted in addressing such issues to ensure models don't intentionally or unintentionally perpetuate bias. This survey offers a synopsis of the different ways researchers have promoted fairness in AI systems. We explore the different definitions of fairness existing in the current literature. We create a comprehensive taxonomy by categorizing different types of bias and investigate cases of biased AI in different application domains. A thorough study is conducted of the approaches and techniques employed by researchers to mitigate bias in AI models. Moreover, we also delve into the impact of biased models on user experience and the ethical considerations to contemplate when developing and deploying such models. We hope this survey helps researchers and practitioners understand the intricate details of fairness and bias in AI systems. By sharing this thorough survey, we aim to promote additional discourse in the domain of equitable and responsible AI.

comment: 37 pages, 6 figures

☆ Deep Support Vectors

Junhoo Lee, Hyunho Lee, Kyomin Hwang, Nojun Kwak

While the success of deep learning is commonly attributed to its theoretical equivalence with Support Vector Machines (SVM), the practical implications of this relationship have not been thoroughly explored. This paper pioneers an exploration in this domain, specifically focusing on the identification of Deep Support Vectors (DSVs) within deep learning models. We introduce the concept of DeepKKT conditions, an adaptation of the traditional Karush-Kuhn-Tucker (KKT) conditions tailored for deep learning. Through empirical investigations, we illustrate that DSVs exhibit similarities to support vectors in SVM, offering a tangible method to interpret the decision-making criteria of models. Additionally, our findings demonstrate that models can be effectively reconstructed using DSVs, resembling the process in SVM. The code will be available.

☆ ALISA: Accelerating Large Language Model Inference via Sparsity-Aware KV Caching ISCA 2024

Youpeng Zhao, Di Wu, Jun Wang

The Transformer architecture has significantly advanced natural language processing (NLP) and has been foundational in developing large language models (LLMs) such as LLaMA and OPT, which have come to dominate a broad range of NLP tasks. Despite their superior accuracy, LLMs present unique challenges in practical inference, concerning the compute and memory-intensive nature. Thanks to the autoregressive characteristic of LLM inference, KV caching for the attention layers in Transformers can effectively accelerate LLM inference by substituting quadratic-complexity computation with linear-complexity memory accesses. Yet, this approach requires increasing memory as demand grows for processing longer sequences. The overhead leads to reduced throughput due to I/O bottlenecks and even out-of-memory errors, particularly on resource-constrained systems like a single commodity GPU. In this paper, we propose ALISA, a novel algorithm-system co-design solution to address the challenges imposed by KV caching. On the algorithm level, ALISA prioritizes tokens that are most important in generating a new token via a Sparse Window Attention (SWA) algorithm. SWA introduces high sparsity in attention layers and reduces the memory footprint of KV caching at negligible accuracy loss. On the system level, ALISA employs three-phase token-level dynamical scheduling and optimizes the trade-off between caching and recomputation, thus maximizing the overall performance in resource-constrained systems. In a single GPU-CPU system, we demonstrate that under varying workloads, ALISA improves the throughput of baseline systems such as FlexGen and vLLM by up to 3X and 1.9X, respectively.

comment: ISCA 2024

☆ Neural Multimodal Topic Modeling: A Comprehensive Evaluation LREC

Felipe González-Pizarro, Giuseppe Carenini

Neural topic models can successfully find coherent and diverse topics in textual data. However, they are limited in dealing with multimodal datasets (e.g., images and text). This paper presents the first systematic and comprehensive evaluation of multimodal topic modeling of documents containing both text and images. In the process, we propose two novel topic modeling solutions and two novel evaluation metrics. Overall, our evaluation on an unprecedented rich and diverse collection of datasets indicates that both of our models generate coherent and diverse topics. Nevertheless, the extent to which one method outperforms the other depends on the metrics and dataset combinations, which suggests further exploration of hybrid solutions in the future. Notably, our succinct human evaluation aligns with the outcomes determined by our proposed metrics. This alignment not only reinforces the credibility of our metrics but also highlights the potential for their application in guiding future multimodal topic modeling endeavors.

comment: Camera-Ready for LREC-COLING 2024 (Long Paper)

☆ Hawk: Accurate and Fast Privacy-Preserving Machine Learning Using Secure Lookup Table Computation

Hamza Saleem, Amir Ziashahabi, Muhammad Naveed, Salman Avestimehr

Training machine learning models on data from multiple entities without direct data sharing can unlock applications otherwise hindered by business, legal, or ethical constraints. In this work, we design and implement new privacy-preserving machine learning protocols for logistic regression and neural network models. We adopt a two-server model where data owners secret-share their data between two servers that train and evaluate the model on the joint data. A significant source of inefficiency and inaccuracy in existing methods arises from using Yao's garbled circuits to compute non-linear activation functions. We propose new methods for computing non-linear functions based on secret-shared lookup tables, offering both computational efficiency and improved accuracy. Beyond introducing leakage-free techniques, we initiate the exploration of relaxed security measures for privacy-preserving machine learning. Instead of claiming that the servers gain no knowledge during the computation, we contend that while some information is revealed about access patterns to lookup tables, it maintains epsilon-dX-privacy. Leveraging this relaxation significantly reduces the computational resources needed for training. We present new cryptographic protocols tailored to this relaxed security paradigm and define and analyze the leakage. Our evaluations show that our logistic regression protocol is up to 9x faster, and the neural network training is up to 688x faster than SecureML. Notably, our neural network achieves an accuracy of 96.6% on MNIST in 15 epochs, outperforming prior benchmarks that capped at 93.4% using the same architecture.

comment: Accepted at Privacy Enhancing Technologies Symposium (PETS) 2024

☆ Not All Federated Learning Algorithms Are Created Equal: A Performance Evaluation Study

Gustav A. Baumgart, Jaemin Shin, Ali Payani, Myungjin Lee, Ramana Rao Kompella

Federated Learning (FL) emerged as a practical approach to training a model from decentralized data. The proliferation of FL led to the development of numerous FL algorithms and mechanisms. Many prior efforts have given their primary focus on accuracy of those approaches, but there exists little understanding of other aspects such as computational overheads, performance and training stability, etc. To bridge this gap, we conduct extensive performance evaluation on several canonical FL algorithms (FedAvg, FedProx, FedYogi, FedAdam, SCAFFOLD, and FedDyn) by leveraging an open-source federated learning framework called Flame. Our comprehensive measurement study reveals that no single algorithm works best across different performance metrics. A few key observations are: (1) While some state-of-the-art algorithms achieve higher accuracy than others, they incur either higher computation overheads (FedDyn) or communication overheads (SCAFFOLD). (2) Recent algorithms present smaller standard deviation in accuracy across clients than FedAvg, indicating that the advanced algorithms' performances are stable. (3) However, algorithms such as FedDyn and SCAFFOLD are more prone to catastrophic failures without the support of additional techniques such as gradient clipping. We hope that our empirical study can help the community to build best practices in evaluating FL algorithms.

☆ An Analysis of Switchback Designs in Reinforcement Learning

Qianglin Wen, Chengchun Shi, Ying Yang, Niansheng Tang, Hongtu Zhu

This paper offers a detailed investigation of switchback designs in A/B testing, which alternate between baseline and new policies over time. Our aim is to thoroughly evaluate the effects of these designs on the accuracy of their resulting average treatment effect (ATE) estimators. We propose a novel "weak signal analysis" framework, which substantially simplifies the calculations of the mean squared errors (MSEs) of these ATEs in Markov decision process environments. Our findings suggest that (i) when the majority of reward errors are positively correlated, the switchback design is more efficient than the alternating-day design which switches policies in a daily basis. Additionally, increasing the frequency of policy switches tends to reduce the MSE of the ATE estimator. (ii) When the errors are uncorrelated, however, all these designs become asymptotically equivalent. (iii) In cases where the majority of errors are negative correlated, the alternating-day design becomes the optimal choice. These insights are crucial, offering guidelines for practitioners on designing experiments in A/B testing. Our analysis accommodates a variety of policy value estimators, including model-based estimators, least squares temporal difference learning estimators, and double reinforcement learning estimators, thereby offering a comprehensive understanding of optimal design strategies for policy evaluation in reinforcement learning.

♻ ☆ Enhancing Zero-Shot Chain-of-Thought Reasoning in Large Language Models through Logic COLING 2024

Xufeng Zhao, Mengdi Li, Wenhao Lu, Cornelius Weber, Jae Hee Lee, Kun Chu, Stefan Wermter

Recent advancements in large language models have showcased their remarkable generalizability across various domains. However, their reasoning abilities still have significant room for improvement, especially when confronted with scenarios requiring multi-step reasoning. Although large language models possess extensive knowledge, their reasoning often fails to effectively utilize this knowledge to establish a coherent thinking paradigm. These models sometimes show hallucinations as their reasoning procedures are unconstrained by logical principles. Aiming at improving the zero-shot chain-of-thought reasoning ability of large language models, we propose LoT (Logical Thoughts), a self-improvement prompting framework that leverages principles rooted in symbolic logic, particularly Reductio ad Absurdum, to systematically verify and rectify the reasoning processes step by step. Experimental evaluations conducted on language tasks in diverse domains, including arithmetic, commonsense, symbolic, causal inference, and social problems, demonstrate the efficacy of enhanced reasoning by logic. The implementation code for LoT can be accessed at: https://github.com/xf-zhao/LoT.

comment: Accepted in COLING 2024. Code see https://github.com/xf-zhao/LoT

♻ ☆ LocalTweets to LocalHealth: A Mental Health Surveillance Framework Based on Twitter Data

Vijeta Deshpande, Minhwa Lee, Zonghai Yao, Zihao Zhang, Jason Brian Gibbons, Hong Yu

Prior research on Twitter (now X) data has provided positive evidence of its utility in developing supplementary health surveillance systems. In this study, we present a new framework to surveil public health, focusing on mental health (MH) outcomes. We hypothesize that locally posted tweets are indicative of local MH outcomes and collect tweets posted from 765 neighborhoods (census block groups) in the USA. We pair these tweets from each neighborhood with the corresponding MH outcome reported by the Center for Disease Control (CDC) to create a benchmark dataset, LocalTweets. With LocalTweets, we present the first population-level evaluation task for Twitter-based MH surveillance systems. We then develop an efficient and effective method, LocalHealth, for predicting MH outcomes based on LocalTweets. When used with GPT3.5, LocalHealth achieves the highest F1-score and accuracy of 0.7429 and 79.78\%, respectively, a 59\% improvement in F1-score over the GPT3.5 in zero-shot setting. We also utilize LocalHealth to extrapolate CDC's estimates to proxy unreported neighborhoods, achieving an F1-score of 0.7291. Our work suggests that Twitter data can be effectively leveraged to simulate neighborhood-level MH outcomes.

♻ ☆ Simple and Scalable Strategies to Continually Pre-train Large Language Models

Adam Ibrahim, Benjamin Thérien, Kshitij Gupta, Mats L. Richter, Quentin Anthony, Timothée Lesort, Eugene Belilovsky, Irina Rish

Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution is to continually pre-train these models, saving significant compute compared to re-training. However, the distribution shift induced by new data typically results in degraded performance on previous data or poor adaptation to the new data. In this work, we show that a simple and scalable combination of learning rate (LR) re-warming, LR re-decaying, and replay of previous data is sufficient to match the performance of fully re-training from scratch on all available data, as measured by the final loss and the average score on several language model (LM) evaluation benchmarks. Specifically, we show this for a weak but realistic distribution shift between two commonly used LLM pre-training datasets (English$\rightarrow$English) and a stronger distribution shift (English$\rightarrow$German) at the $405$M parameter model scale with large dataset sizes (hundreds of billions of tokens). Selecting the weak but realistic shift for larger-scale experiments, we also find that our continual learning strategies match the re-training baseline for a 10B parameter LLM. Our results demonstrate that LLMs can be successfully updated via simple and scalable continual learning strategies, matching the re-training baseline using only a fraction of the compute. Finally, inspired by previous work, we propose alternatives to the cosine learning rate schedule that help circumvent forgetting induced by LR re-warming and that are not bound to a fixed token budget.

♻ ☆ An optimal control perspective on diffusion-based generative modeling NeurIPS 2022

Julius Berner, Lorenz Richter, Karen Ullrich

We establish a connection between stochastic optimal control and generative models based on stochastic differential equations (SDEs), such as recently developed diffusion probabilistic models. In particular, we derive a Hamilton-Jacobi-Bellman equation that governs the evolution of the log-densities of the underlying SDE marginals. This perspective allows to transfer methods from optimal control theory to generative modeling. First, we show that the evidence lower bound is a direct consequence of the well-known verification theorem from control theory. Further, we can formulate diffusion-based generative modeling as a minimization of the Kullback-Leibler divergence between suitable measures in path space. Finally, we develop a novel diffusion-based method for sampling from unnormalized densities -- a problem frequently occurring in statistics and computational sciences. We demonstrate that our time-reversed diffusion sampler (DIS) can outperform other diffusion-based sampling approaches on multiple numerical examples.

comment: Accepted for oral presentation at NeurIPS 2022 Workshop on Score-Based Methods

♻ ☆ Fully Independent Communication in Multi-Agent Reinforcement Learning AAMAS 2024

Rafael Pina, Varuna De Silva, Corentin Artaud, Xiaolan Liu

Multi-Agent Reinforcement Learning (MARL) comprises a broad area of research within the field of multi-agent systems. Several recent works have focused specifically on the study of communication approaches in MARL. While multiple communication methods have been proposed, these might still be too complex and not easily transferable to more practical contexts. One of the reasons for that is due to the use of the famous parameter sharing trick. In this paper, we investigate how independent learners in MARL that do not share parameters can communicate. We demonstrate that this setting might incur into some problems, to which we propose a new learning scheme as a solution. Our results show that, despite the challenges, independent agents can still learn communication strategies following our method. Additionally, we use this method to investigate how communication in MARL is affected by different network capacities, both for sharing and not sharing parameters. We observe that communication may not always be needed and that the chosen agent network sizes need to be considered when used together with communication in order to achieve efficient learning.

comment: Extended version of the paper appearing on AAMAS 2024 with the same title. 11 pages, 8 figures

♻ ☆ A randomized algorithm for nonconvex minimization with inexact evaluations and complexity guarantees

Shuyao Li, Stephen J. Wright

We consider minimization of a smooth nonconvex function with inexact oracle access to gradient and Hessian (without assuming access to the function value) to achieve approximate second-order optimality. A novel feature of our method is that if an approximate direction of negative curvature is chosen as the step, we choose its sense to be positive or negative with equal probability. We allow gradients to be inexact in a relative sense and relax the coupling between inexactness thresholds for the first- and second-order optimality conditions. Our convergence analysis includes both an expectation bound based on martingale analysis and a high-probability bound based on concentration inequalities. We apply our algorithm to empirical risk minimization problems and obtain improved gradient sample complexity over existing works.

♻ ☆ Borrowing Treasures from Neighbors: In-Context Learning for Multimodal Learning with Missing Modalities and Data Scarcity

Zhuo Zhi, Ziquan Liu, Moe Elbadawi, Adam Daneshmend, Mine Orlu, Abdul Basit, Andreas Demosthenous, Miguel Rodrigues

Multimodal machine learning with missing modalities is an increasingly relevant challenge arising in various applications such as healthcare. This paper extends the current research into missing modalities to the low-data regime, i.e., a downstream task has both missing modalities and limited sample size issues. This problem setting is particularly challenging and also practical as it is often expensive to get full-modality data and sufficient annotated training samples. We propose to use retrieval-augmented in-context learning to address these two crucial issues by unleashing the potential of a transformer's in-context learning ability. Diverging from existing methods, which primarily belong to the parametric paradigm and often require sufficient training samples, our work exploits the value of the available full-modality data, offering a novel perspective on resolving the challenge. The proposed data-dependent framework exhibits a higher degree of sample efficiency and is empirically demonstrated to enhance the classification model's performance on both full- and missing-modality data in the low-data regime across various multimodal learning tasks. When only 1% of the training data are available, our proposed method demonstrates an average improvement of 6.1% over a recent strong baseline across various datasets and missing states. Notably, our method also reduces the performance gap between full-modality and missing-modality data compared with the baseline.

♻ ☆ Probabilistically Rewired Message-Passing Neural Networks ICLR 2024

Chendi Qian, Andrei Manolache, Kareem Ahmed, Zhe Zeng, Guy Van den Broeck, Mathias Niepert, Christopher Morris

Message-passing graph neural networks (MPNNs) emerged as powerful tools for processing graph-structured input. However, they operate on a fixed input graph structure, ignoring potential noise and missing information. Furthermore, their local aggregation mechanism can lead to problems such as over-squashing and limited expressive power in capturing relevant graph structures. Existing solutions to these challenges have primarily relied on heuristic methods, often disregarding the underlying data distribution. Hence, devising principled approaches for learning to infer graph structures relevant to the given prediction task remains an open challenge. In this work, leveraging recent progress in exact and differentiable $k$-subset sampling, we devise probabilistically rewired MPNNs (PR-MPNNs), which learn to add relevant edges while omitting less beneficial ones. For the first time, our theoretical analysis explores how PR-MPNNs enhance expressive power, and we identify precise conditions under which they outperform purely randomized approaches. Empirically, we demonstrate that our approach effectively mitigates issues like over-squashing and under-reaching. In addition, on established real-world datasets, our method exhibits competitive or superior predictive performance compared to traditional MPNN models and recent graph transformer architectures.

comment: ICLR 2024

♻ ☆ Optimal Data Splitting in Distributed Optimization for Machine Learning

Daniil Medyakov, Gleb Molodtsov, Aleksandr Beznosikov, Alexander Gasnikov

The distributed optimization problem has become increasingly relevant recently. It has a lot of advantages such as processing a large amount of data in less time compared to non-distributed methods. However, most distributed approaches suffer from a significant bottleneck - the cost of communications. Therefore, a large amount of research has recently been directed at solving this problem. One such approach uses local data similarity. In particular, there exists an algorithm provably optimally exploiting the similarity property. But this result, as well as results from other works solve the communication bottleneck by focusing only on the fact that communication is significantly more expensive than local computing and does not take into account the various capacities of network devices and the different relationship between communication time and local computing expenses. We consider this setup and the objective of this study is to achieve an optimal ratio of distributed data between the server and local machines for any costs of communications and local computations. The running times of the network are compared between uniform and optimal distributions. The superior theoretical performance of our solutions is experimentally validated.

comment: 17 pages, 2 figures

♻ ☆ Dynamics of Moral Behavior in Heterogeneous Populations of Learning Agents

Elizaveta Tennant, Stephen Hailes, Mirco Musolesi

Growing concerns about safety and alignment of AI systems highlight the importance of embedding moral capabilities in artificial agents. A promising solution is the use of learning from experience, i.e., Reinforcement Learning. In multi-agent (social) environments, complex population-level phenomena may emerge from interactions between individual learning agents. Many of the existing studies rely on simulated social dilemma environments to study the interactions of independent learning agents. However, they tend to ignore the moral heterogeneity that is likely to be present in societies of agents in practice. For example, at different points in time a single learning agent may face opponents who are consequentialist (i.e., caring about maximizing some outcome over time) or norm-based (i.e., focusing on conforming to a specific norm here and now). The extent to which agents' co-development may be impacted by such moral heterogeneity in populations is not well understood. In this paper, we present a study of the learning dynamics of morally heterogeneous populations interacting in a social dilemma setting. Using a Prisoner's Dilemma environment with a partner selection mechanism, we investigate the extent to which the prevalence of diverse moral agents in populations affects individual agents' learning behaviors and emergent population-level outcomes. We observe several types of non-trivial interactions between pro-social and anti-social agents, and find that certain classes of moral agents are able to steer selfish agents towards more cooperative behavior.

♻ ☆ Room Transfer Function Reconstruction Using Complex-valued Neural Networks and Irregularly Distributed Microphones

Francesca Ronchini, Luca Comanducci, Mirco Pezzoli, Fabio Antonacci, Augusto Sarti

Reconstructing the room transfer functions needed to calculate the complex sound field in a room has several impor- tant real-world applications. However, an unpractical number of microphones is often required. Recently, in addition to classical signal processing methods, deep learning techniques have been applied to reconstruct the room transfer function starting from a very limited set of measurements at scattered points in the room. In this paper, we employ complex-valued neural networks to estimate room transfer functions in the frequency range of the first room resonances, using a few irregularly distributed microphones. To the best of our knowledge, this is the first time that complex-valued neural networks are used to estimate room transfer functions. To analyze the benefits of applying complex- valued optimization to the considered task, we compare the proposed technique with a state-of-the-art kernel-based signal processing approach for sound field reconstruction, showing that the proposed technique exhibits relevant advantages in terms of phase accuracy and overall quality of the reconstructed sound field. For informative purposes, we also compare the model with a similarly-structured data-driven approach that, however, applies a real-valued neural network to reconstruct only the magnitude of the sound field.

comment: Submitted to EUSIPCO 2024

♻ ☆ Activations and Gradients Compression for Model-Parallel Training

Mikhail Rudakov, Aleksandr Beznosikov, Yaroslav Kholodov, Alexander Gasnikov

Large neural networks require enormous computational clusters of machines. Model-parallel training, when the model architecture is partitioned sequentially between workers, is a popular approach for training modern models. Information compression can be applied to decrease workers communication time, as it is often a bottleneck in such systems. This work explores how simultaneous compression of activations and gradients in model-parallel distributed training setup affects convergence. We analyze compression methods such as quantization and TopK compression, and also experiment with error compensation techniques. Moreover, we employ TopK with AQ-SGD per-batch error feedback approach. We conduct experiments on image classification and language model fine-tuning tasks. Our findings demonstrate that gradients require milder compression rates than activations. We observe that $K=10\%$ is the lowest TopK compression level, which does not harm model convergence severely. Experiments also show that models trained with TopK perform well only when compression is also applied during inference. We find that error feedback techniques do not improve model-parallel training compared to plain compression, but allow model inference without compression with almost no quality drop. Finally, when applied with the AQ-SGD approach, TopK stronger than with $ K=30\%$ worsens model performance significantly.

comment: 17 pages, 6 figures, 5 tables

♻ ☆ Differentially private multivariate medians

Kelly Ramsay, Aukosh Jagannath, Shoja'eddin Chenouri

Statistical tools which satisfy rigorous privacy guarantees are necessary for modern data analysis. It is well-known that robustness against contamination is linked to differential privacy. Despite this fact, using multivariate medians for differentially private and robust multivariate location estimation has not been systematically studied. We develop novel finite-sample performance guarantees for differentially private multivariate depth-based medians, which are essentially sharp. Our results cover commonly used depth functions, such as the halfspace (or Tukey) depth, spatial depth, and the integrated dual depth. We show that under Cauchy marginals, the cost of heavy-tailed location estimation outweighs the cost of privacy. We demonstrate our results numerically using a Gaussian contamination model in dimensions up to d = 100, and compare them to a state-of-the-art private mean estimation algorithm. As a by-product of our investigation, we prove concentration inequalities for the output of the exponential mechanism about the maximizer of the population objective function. This bound applies to objective functions that satisfy a mild regularity condition.

comment: 42 pages, 3 figures, 2 tables

♻ ☆ AI and Generative AI for Research Discovery and Summarization

Mark Glickman, Yi Zhang

AI and generative AI tools, including chatbots like ChatGPT that rely on large language models (LLMs), have burst onto the scene this year, creating incredible opportunities to increase work productivity and improve our lives. Statisticians and data scientists have begun experiencing the benefits from the availability of these tools in numerous ways, such as the generation of programming code from text prompts to analyze data or fit statistical models. One area that these tools can make a substantial impact is in research discovery and summarization. Standalone tools and plugins to chatbots are being developed that allow researchers to more quickly find relevant literature than pre-2023 search tools. Furthermore, generative AI tools have improved to the point where they can summarize and extract the key points from research articles in succinct language. Finally, chatbots based on highly parameterized LLMs can be used to simulate abductive reasoning, which provides researchers the ability to make connections among related technical topics, which can also be used for research discovery. We review the developments in AI and generative AI for research discovery and summarization, and propose directions where these types of tools are likely to head in the future that may be of interest to statistician and data scientists.

comment: 29 pages, 9 figures

♻ ☆ PINN surrogate of Li-ion battery models for parameter inference. Part II: Regularization and application of the pseudo-2D model

Malik Hassanaly, Peter J. Weddle, Ryan N. King, Subhayan De, Alireza Doostan, Corey R. Randall, Eric J. Dufek, Andrew M. Colclasure, Kandler Smith

Bayesian parameter inference is useful to improve Li-ion battery diagnostics and can help formulate battery aging models. However, it is computationally intensive and cannot be easily repeated for multiple cycles, multiple operating conditions, or multiple replicate cells. To reduce the computational cost of Bayesian calibration, numerical solvers for physics-based models can be replaced with faster surrogates. A physics-informed neural network (PINN) is developed as a surrogate for the pseudo-2D (P2D) battery model calibration. For the P2D surrogate, additional training regularization was needed as compared to the PINN single-particle model (SPM) developed in Part I. Both the PINN SPM and P2D surrogate models are exercised for parameter inference and compared to data obtained from a direct numerical solution of the governing equations. A parameter inference study highlights the ability to use these PINNs to calibrate scaling parameters for the cathode Li diffusion and the anode exchange current density. By realizing computational speed-ups of 2250x for the P2D model, as compared to using standard integrating methods, the PINN surrogates enable rapid state-of-health diagnostics. In the low-data availability scenario, the testing error was estimated to 2mV for the SPM surrogate and 10mV for the P2D surrogate which could be mitigated with additional data.

♻ ☆ ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review

Sunder Ali Khowaja, Parus Khuwaja, Kapal Dev, Weizheng Wang, Lewis Nkenyereye

ChatGPT is another large language model (LLM) vastly available for the consumers on their devices but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail the issues and concerns raised over chatGPT in line with aforementioned characteristics. We also discuss the recent EU AI Act briefly in accordance with the SPADE evaluation. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for AI policy act, if designed by the governments.

comment: 29 pages, 8 figures, 4 tables

♻ ☆ PINN surrogate of Li-ion battery models for parameter inference. Part I: Implementation and multi-fidelity hierarchies for the single-particle model

Malik Hassanaly, Peter J. Weddle, Ryan N. King, Subhayan De, Alireza Doostan, Corey R. Randall, Eric J. Dufek, Andrew M. Colclasure, Kandler Smith

To plan and optimize energy storage demands that account for Li-ion battery aging dynamics, techniques need to be developed to diagnose battery internal states accurately and rapidly. This study seeks to reduce the computational resources needed to determine a battery's internal states by replacing physics-based Li-ion battery models -- such as the single-particle model (SPM) and the pseudo-2D (P2D) model -- with a physics-informed neural network (PINN) surrogate. The surrogate model makes high-throughput techniques, such as Bayesian calibration, tractable to determine battery internal parameters from voltage responses. This manuscript is the first of a two-part series that introduces PINN surrogates of Li-ion battery models for parameter inference (i.e., state-of-health diagnostics). In this first part, a method is presented for constructing a PINN surrogate of the SPM. A multi-fidelity hierarchical training, where several neural nets are trained with multiple physics-loss fidelities is shown to significantly improve the surrogate accuracy when only training on the governing equation residuals. The implementation is made available in a companion repository (https://github.com/NREL/pinnstripes). The techniques used to develop a PINN surrogate of the SPM are extended in Part II for the PINN surrogate for the P2D battery model, and explore the Bayesian calibration capabilities of both surrogates.

♻ ☆ Disentangling the Spectral Properties of the Hodge Laplacian: Not All Small Eigenvalues Are Equal

Vincent P. Grande, Michael T. Schaub

The rich spectral information of the graph Laplacian has been instrumental in graph theory, machine learning, and graph signal processing for applications such as graph classification, clustering, or eigenmode analysis. Recently, the Hodge Laplacian has come into focus as a generalisation of the ordinary Laplacian for higher-order graph models such as simplicial and cellular complexes. Akin to the traditional analysis of graph Laplacians, many authors analyse the smallest eigenvalues of the Hodge Laplacian, which are connected to important topological properties such as homology. However, small eigenvalues of the Hodge Laplacian can carry different information depending on whether they are related to curl or gradient eigenmodes, and thus may not be comparable. We therefore introduce the notion of persistent eigenvector similarity and provide a method to track individual harmonic, curl, and gradient eigenvectors/-values through the so-called persistence filtration, leveraging the full information contained in the Hodge-Laplacian spectrum across all possible scales of a point cloud. Finally, we use our insights (a) to introduce a novel form of Hodge spectral clustering and (b) to classify edges and higher-order simplices based on their relationship to the smallest harmonic, curl, and gradient eigenvectors.

comment: 5 pages, 4 figures, comments welcome

♻ ☆ Semi-Supervised Crowd Counting from Unlabeled Data

Haoran Duan, Fan Wan, Rui Sun, Zeyu Wang, Varun Ojha, Yu Guan, Hubert P. H. Shum, Bingzhang Hu, Yang Long

Automatic Crowd behavior analysis can be applied to effectively help the daily transportation statistics and planning, which helps the smart city construction. As one of the most important keys, crowd counting has drawn increasing attention. Recent works achieved promising performance but relied on the supervised paradigm with expensive crowd annotations. To alleviate the annotation cost in real-world transportation scenarios, in this work we proposed a semi-supervised learning framework $S^{4}\textit{Crowd}$, which can leverage both unlabeled/labeled data for robust crowd counting. In the unsupervised pathway, two \textit{self-supervised losses} were proposed to simulate the crowd variations such as scale, illumination, based on which supervised information pseudo labels were generated and gradually refined. We also proposed a crowd-driven recurrent unit \textit{Gated-Crowd-Recurrent-Unit (GCRU)}, which can preserve discriminant crowd information by extracting second-order statistics, yielding pseudo labels with improved quality. A joint loss including both unsupervised/supervised information was proposed, and a dynamic weighting strategy was employed to balance the importance of the unsupervised loss and supervised loss at different training stages. We conducted extensive experiments on four popular crowd counting datasets in semi-supervised settings. Experimental results supported the effectiveness of each proposed component in our $S^{4}$Crowd framework. Our method achieved competitive performance in semi-supervised learning approaches on these crowd counting datasets.

♻ ☆ Efficient Pre-training for Localized Instruction Generation of Videos

Anil Batra, Davide Moltisanti, Laura Sevilla-Lara, Marcus Rohrbach, Frank Keller

Procedural videos show step-by-step demonstrations of tasks like recipe preparation. Understanding such videos is challenging, involving the precise localization of steps and the generation of textual instructions. Manually annotating steps and writing instructions is costly, which limits the size of current datasets and hinders effective learning. Leveraging large but noisy video-transcript datasets for pre-training can boost performance, but demands significant computational resources. Furthermore, transcripts contain irrelevant content and exhibit style variation compared to instructions written by human annotators. To mitigate both issues, we propose a technique, Sieve-&-Swap, to automatically curate a smaller dataset: (i) Sieve filters irrelevant transcripts and (ii) Swap enhances the quality of the text instruction by automatically replacing the transcripts with human-written instructions from a text-only recipe dataset. The curated dataset, three orders of magnitude smaller than current web-scale datasets, enables efficient training of large-scale models with competitive performance. We complement our Sieve-\&-Swap approach with a Procedure Transformer (ProcX) for end-to-end step localization and instruction generation for procedural videos. When this model is pre-trained on our curated dataset, it achieves state-of-the-art performance in zero-shot and finetuning settings on YouCook2 and Tasty, while using a fraction of the computational resources.

comment: This version has some missing experiments and elaborative technical details

♻ ☆ Toward a Theory of Causation for Interpreting Neural Code Models

David N. Palacio, Alejandro Velasco, Nathan Cooper, Alvaro Rodriguez, Kevin Moran, Denys Poshyvanyk

Neural Language Models of Code, or Neural Code Models (NCMs), are rapidly progressing from research prototypes to commercial developer tools. As such, understanding the capabilities and limitations of such models is becoming critical. However, the abilities of these models are typically measured using automated metrics that often only reveal a portion of their real-world performance. While, in general, the performance of NCMs appears promising, currently much is unknown about how such models arrive at decisions. To this end, this paper introduces $do_{code}$, a post hoc interpretability method specific to NCMs that is capable of explaining model predictions. $do_{code}$ is based upon causal inference to enable programming language-oriented explanations. While the theoretical underpinnings of $do_{code}$ are extensible to exploring different model properties, we provide a concrete instantiation that aims to mitigate the impact of spurious correlations by grounding explanations of model behavior in properties of programming languages. To demonstrate the practical benefit of $do_{code}$, we illustrate the insights that our framework can provide by performing a case study on two popular deep learning architectures and ten NCMs. The results of this case study illustrate that our studied NCMs are sensitive to changes in code syntax. All our NCMs, except for the BERT-like model, statistically learn to predict tokens related to blocks of code (\eg brackets, parenthesis, semicolon) with less confounding bias as compared to other programming language constructs. These insights demonstrate the potential of $do_{code}$ as a useful method to detect and facilitate the elimination of confounding bias in NCMs.

comment: Accepted to appear in IEEE Transactions on Software Engineering

♻ ☆ RL$^3$: Boosting Meta Reinforcement Learning via RL inside RL$^2$

Abhinav Bhatia, Samer B. Nashed, Shlomo Zilberstein

Meta reinforcement learning (meta-RL) methods such as RL$^2$ have emerged as promising approaches for learning data-efficient RL algorithms tailored to a given task distribution. However, they show poor asymptotic performance and struggle with out-of-distribution tasks because they rely on sequence models, such as recurrent neural networks or transformers, to process experiences rather than summarize them using general-purpose RL components such as value functions. In contrast, traditional RL algorithms are data-inefficient as they do not use domain knowledge, but they do converge to an optimal policy in the limit. We propose RL$^3$, a principled hybrid approach that incorporates action-values, learned per task through traditional RL, in the inputs to meta-RL. We show that RL$^3$ earns greater cumulative reward in the long term, compared to RL$^2$, while maintaining data-efficiency in the short term, and generalizes better to out-of-distribution tasks. Experiments are conducted on both custom and benchmark discrete domains from the meta-RL literature that exhibit a range of short-term, long-term, and complex dependencies.

♻ ☆ Multi-Objective Optimization for Sparse Deep Multi-Task Learning

S. S. Hotegni, M. Berkemeier, S. Peitz

Different conflicting optimization criteria arise naturally in various Deep Learning scenarios. These can address different main tasks (i.e., in the setting of Multi-Task Learning), but also main and secondary tasks such as loss minimization versus sparsity. The usual approach is a simple weighting of the criteria, which formally only works in the convex setting. In this paper, we present a Multi-Objective Optimization algorithm using a modified Weighted Chebyshev scalarization for training Deep Neural Networks (DNNs) with respect to several tasks. By employing this scalarization technique, the algorithm can identify all optimal solutions of the original problem while reducing its complexity to a sequence of single-objective problems. The simplified problems are then solved using an Augmented Lagrangian method, enabling the use of popular optimization techniques such as Adam and Stochastic Gradient Descent, while efficaciously handling constraints. Our work aims to address the (economical and also ecological) sustainability issue of DNN models, with a particular focus on Deep Multi-Task models, which are typically designed with a very large number of weights to perform equally well on multiple tasks. Through experiments conducted on two Machine Learning datasets, we demonstrate the possibility of adaptively sparsifying the model during training without significantly impacting its performance, if we are willing to apply task-specific adaptations to the network weights. Code is available at https://github.com/salomonhotegni/MDMTN

comment: 12 pages, 7 figures

♻ ☆ Accelerating Scientific Discovery with Generative Knowledge Extraction, Graph-Based Representation, and Multimodal Intelligent Graph Reasoning

Markus J. Buehler

Leveraging generative Artificial Intelligence (AI), we have transformed a dataset comprising 1,000 scientific papers into an ontological knowledge graph. Through an in-depth structural analysis, we have calculated node degrees, identified communities and connectivities, and evaluated clustering coefficients and betweenness centrality of pivotal nodes, uncovering fascinating knowledge architectures. The graph has an inherently scale-free nature, is highly connected, and can be used for graph reasoning by taking advantage of transitive and isomorphic properties that reveal unprecedented interdisciplinary relationships that can be used to answer queries, identify gaps in knowledge, propose never-before-seen material designs, and predict material behaviors. We compute deep node embeddings for combinatorial node similarity ranking for use in a path sampling strategy links dissimilar concepts that have previously not been related. One comparison revealed structural parallels between biological materials and Beethoven's 9th Symphony, highlighting shared patterns of complexity through isomorphic mapping. In another example, the algorithm proposed a hierarchical mycelium-based composite based on integrating path sampling with principles extracted from Kandinsky's 'Composition VII' painting. The resulting material integrates an innovative set of concepts that include a balance of chaos/order, adjustable porosity, mechanical strength, and complex patterned chemical functionalization. We uncover other isomorphisms across science, technology and art, revealing a nuanced ontology of immanence that reveal a context-dependent heterarchical interplay of constituents. Graph-based generative AI achieves a far higher degree of novelty, explorative capacity, and technical detail, than conventional approaches and establishes a widely useful framework for innovation by revealing hidden connections.

♻ ☆ Unveiling the Pitfalls of Knowledge Editing for Large Language Models ICLR 2024

Zhoubo Li, Ningyu Zhang, Yunzhi Yao, Mengru Wang, Xi Chen, Huajun Chen

As the cost associated with fine-tuning Large Language Models (LLMs) continues to rise, recent research efforts have pivoted towards developing methodologies to edit implicit knowledge embedded within LLMs. Yet, there's still a dark cloud lingering overhead -- will knowledge editing trigger butterfly effect? since it is still unclear whether knowledge editing might introduce side effects that pose potential risks or not. This paper pioneers the investigation into the potential pitfalls associated with knowledge editing for LLMs. To achieve this, we introduce new benchmark datasets and propose innovative evaluation metrics. Our results underline two pivotal concerns: (1) Knowledge Conflict: Editing groups of facts that logically clash can magnify the inherent inconsistencies in LLMs-a facet neglected by previous methods. (2) Knowledge Distortion: Altering parameters with the aim of editing factual knowledge can irrevocably warp the innate knowledge structure of LLMs. Experimental results vividly demonstrate that knowledge editing might inadvertently cast a shadow of unintended consequences on LLMs, which warrant attention and efforts for future works. Code and data are available at https://github.com/zjunlp/PitfallsKnowledgeEditing.

comment: ICLR 2024

♻ ☆ Artificial Neural Nets and the Representation of Human Concepts

Timo Freiesleben

What do artificial neural networks (ANNs) learn? The machine learning (ML) community shares the narrative that ANNs must develop abstract human concepts to perform complex tasks. Some go even further and believe that these concepts are stored in individual units of the network. Based on current research, I systematically investigate the assumptions underlying this narrative. I conclude that ANNs are indeed capable of performing complex prediction tasks, and that they may learn human and non-human concepts to do so. However, evidence indicates that ANNs do not represent these concepts in individual units.

comment: For: Philosophy of Science for Machine Learning: Core Issues and New Perspectives, edited by Juan Duran and Giorgia Pozzi

♻ ☆ Dual Conic Proxies for AC Optimal Power Flow SC

Guancheng Qiu, Mathieu Tanneau, Pascal Van Hentenryck

In recent years, there has been significant interest in the development of machine learning-based optimization proxies for AC Optimal Power Flow (AC-OPF). Although significant progress has been achieved in predicting high-quality primal solutions, no existing learning-based approach can provide valid dual bounds for AC-OPF. This paper addresses this gap by training optimization proxies for a convex relaxation of AC-OPF. Namely, the paper considers a second-order cone (SOC) relaxation of AC-OPF, and proposes \revision{a novel architecture} that embeds a fast, differentiable (dual) feasibility recovery, thus providing valid dual bounds. The paper combines this new architecture with a self-supervised learning scheme, which alleviates the need for costly training data generation. Extensive numerical experiments on medium- and large-scale power grids demonstrate the efficiency and scalability of the proposed methodology.

comment: accepted to PSCC 2024

♻ ☆ Harmonic Control Lyapunov Barrier Functions for Constrained Optimal Control with Reach-Avoid Specifications

Amartya Mukherjee, Ruikun Zhou, Haocheng Chang, Jun Liu

This paper introduces harmonic control Lyapunov barrier functions (harmonic CLBF) that aid in constrained control problems such as reach-avoid problems. Harmonic CLBFs exploit the maximum principle that harmonic functions satisfy to encode the properties of control Lyapunov barrier functions (CLBFs). As a result, they can be initiated at the start of an experiment rather than trained based on sample trajectories. The control inputs are selected to maximize the inner product of the system dynamics with the steepest descent direction of the harmonic CLBF. Numerical results are presented with four different systems under different reach-avoid environments. Harmonic CLBFs show a significantly low risk of entering unsafe regions and a high probability of entering the goal region.

♻ ☆ Investigating Feature and Model Importance in Android Malware Detection: An Implemented Survey and Experimental Comparison of ML-Based Methods

Ali Muzaffar, Hani Ragab Hassen, Hind Zantout, Michael A Lones

The popularity of Android means it is a common target for malware. Over the years, various studies have found that machine learning models can effectively discriminate malware from benign applications. However, as the operating system evolves, so does malware, bringing into question the findings of these previous studies, many of which report very high accuracies using small, outdated, and often imbalanced datasets. In this paper, we reimplement 18 representative past works and reevaluate them using a balanced, relevant, and up-to-date dataset comprising 124,000 applications. We also carry out new experiments designed to fill holes in existing knowledge, and use our findings to identify the most effective features and models to use for Android malware detection within a contemporary environment. We show that high detection accuracies (up to 96.8%) can be achieved using features extracted through static analysis alone, yielding a modest benefit (1%) from using far more expensive dynamic analysis. API calls and opcodes are the most productive static and TCP network traffic provide the most predictive dynamic features. Random forests are generally the most effective model, outperforming more complex deep learning approaches. Whilst directly combining static and dynamic features is generally ineffective, ensembling models separately leads to performances comparable to the best models but using less brittle features.

♻ ☆ In Search of a Data Transformation That Accelerates Neural Field Training CVPR 2024

Junwon Seo, Sangyoon Lee, Kwang In Kim, Jaeho Lee

Neural field is an emerging paradigm in data representation that trains a neural network to approximate the given signal. A key obstacle that prevents its widespread adoption is the encoding speed-generating neural fields requires an overfitting of a neural network, which can take a significant number of SGD steps to reach the desired fidelity level. In this paper, we delve into the impacts of data transformations on the speed of neural field training, specifically focusing on how permuting pixel locations affect the convergence speed of SGD. Counterintuitively, we find that randomly permuting the pixel locations can considerably accelerate the training. To explain this phenomenon, we examine the neural field training through the lens of PSNR curves, loss landscapes, and error patterns. Our analyses suggest that the random pixel permutations remove the easy-to-fit patterns, which facilitate easy optimization in the early stage but hinder capturing fine details of the signal.

comment: CVPR 2024

♻ ☆ Domain Randomization via Entropy Maximization ICLR 2024

Gabriele Tiboni, Pascal Klink, Jan Peters, Tatiana Tommasi, Carlo D'Eramo, Georgia Chalvatzaki

Varying dynamics parameters in simulation is a popular Domain Randomization (DR) approach for overcoming the reality gap in Reinforcement Learning (RL). Nevertheless, DR heavily hinges on the choice of the sampling distribution of the dynamics parameters, since high variability is crucial to regularize the agent's behavior but notoriously leads to overly conservative policies when randomizing excessively. In this paper, we propose a novel approach to address sim-to-real transfer, which automatically shapes dynamics distributions during training in simulation without requiring real-world data. We introduce DOmain RAndomization via Entropy MaximizatiON (DORAEMON), a constrained optimization problem that directly maximizes the entropy of the training distribution while retaining generalization capabilities. In achieving this, DORAEMON gradually increases the diversity of sampled dynamics parameters as long as the probability of success of the current policy is sufficiently high. We empirically validate the consistent benefits of DORAEMON in obtaining highly adaptive and generalizable policies, i.e. solving the task at hand across the widest range of dynamics parameters, as opposed to representative baselines from the DR literature. Notably, we also demonstrate the Sim2Real applicability of DORAEMON through its successful zero-shot transfer in a robotic manipulation setup under unknown real-world parameters.

comment: Published as a conference paper at ICLR 2024. Project website at https://gabrieletiboni.github.io/doraemon/

♻ ☆ Byzantine-resilient Federated Learning With Adaptivity to Data Heterogeneity

Shiyuan Zuo, Xingrun Yan, Rongfei Fan, Han Hu, Hangguan Shan, Tony Q. S. Quek

This paper deals with federated learning (FL) in the presence of malicious Byzantine attacks and data heterogeneity. A novel Robust Average Gradient Algorithm (RAGA) is proposed, which leverages the geometric median for aggregation and can freely select the round number for local updating. Different from most existing resilient approaches, which perform convergence analysis based on strongly-convex loss function or homogeneously distributed dataset, we conduct convergence analysis for not only strongly-convex but also non-convex loss function over heterogeneous dataset. According to our theoretical analysis, as long as the fraction of dataset from malicious users is less than half, RAGA can achieve convergence at rate $\mathcal{O}({1}/{T^{2/3- \delta}})$ where $T$ is the iteration number and $\delta \in (0, 2/3)$ for non-convex loss function, and at linear rate for strongly-convex loss function. Moreover, stationary point or global optimal solution is proved to obtainable as data heterogeneity vanishes. Experimental results corroborate the robustness of RAGA to Byzantine attacks and verifies the advantage of RAGA over baselines on convergence performance under various intensity of Byzantine attacks, for heterogeneous dataset.

♻ ☆ Discretized Distributed Optimization over Dynamic Digraphs

Mohammadreza Doostmohammadian, Wei Jiang, Muwahida Liaquat, Alireza Aghasi, Houman Zarrabi

We consider a discrete-time model of continuous-time distributed optimization over dynamic directed-graphs (digraphs) with applications to distributed learning. Our optimization algorithm works over general strongly connected dynamic networks under switching topologies, e.g., in mobile multi-agent systems and volatile networks due to link failures. Compared to many existing lines of work, there is no need for bi-stochastic weight designs on the links. The existing literature mostly needs the link weights to be stochastic using specific weight-design algorithms needed both at the initialization and at all times when the topology of the network changes. This paper eliminates the need for such algorithms and paves the way for distributed optimization over time-varying digraphs. We derive the bound on the gradient-tracking step-size and discrete time-step for convergence and prove dynamic stability using arguments from consensus algorithms, matrix perturbation theory, and Lyapunov theory. This work, particularly, is an improvement over existing stochastic-weight undirected networks in case of link removal or packet drops. This is because the existing literature may need to rerun time-consuming and computationally complex algorithms for stochastic design, while the proposed strategy works as long as the underlying network is weight-symmetric and balanced. The proposed optimization framework finds applications to distributed classification and learning.

♻ ☆ COPR: Continual Learning Human Preference through Optimal Policy Regularization

Han Zhang, Lin Gui, Yuanzhao Zhai, Hui Wang, Yu Lei, Ruifeng Xu

The technique of Reinforcement Learning from Human Feedback (RLHF) is a commonly employed method to improve pre-trained Language Models (LM), enhancing their ability to conform to human preferences. Nevertheless, the current RLHF-based LMs necessitate full retraining each time novel queries or feedback are introduced, which becomes a challenging task because human preferences can vary between different domains or tasks. Retraining LMs poses practical difficulties in many real-world situations due to the significant time and computational resources required, along with concerns related to data privacy. To address this limitation, we propose a new method called Continual Optimal Policy Regularization (COPR), in which we compute the distribution of optimal policy bypassing the partition function and then regularize the current policy based on the historically optimal distribution to mitigate Catastrophic Forgetting (CF). COPR involves a single learning phase and doesn't necessitate complex reinforcement learning. Importantly, it shares the capability with RLHF to learn from unlabeled data by maintaining a scoring module, similar to reward model, making it flexible for continually learning without human feedback. Our experimental results show that COPR outperforms strong Continuous Learning (CL) baselines when it comes to consistently aligning with human preferences on incremental tasks and domains.

♻ ☆ SD4Match: Learning to Prompt Stable Diffusion Model for Semantic Matching CVPR 2024

Xinghui Li, Jingyi Lu, Kai Han, Victor Prisacariu

In this paper, we address the challenge of matching semantically similar keypoints across image pairs. Existing research indicates that the intermediate output of the UNet within the Stable Diffusion (SD) can serve as robust image feature maps for such a matching task. We demonstrate that by employing a basic prompt tuning technique, the inherent potential of Stable Diffusion can be harnessed, resulting in a significant enhancement in accuracy over previous approaches. We further introduce a novel conditional prompting module that conditions the prompt on the local details of the input image pairs, leading to a further improvement in performance. We designate our approach as SD4Match, short for Stable Diffusion for Semantic Matching. Comprehensive evaluations of SD4Match on the PF-Pascal, PF-Willow, and SPair-71k datasets show that it sets new benchmarks in accuracy across all these datasets. Particularly, SD4Match outperforms the previous state-of-the-art by a margin of 12 percentage points on the challenging SPair-71k dataset.

comment: Accepted to CVPR 2024. Project website: https://sd4match.active.vision/

♻ ☆ Stable Linear Subspace Identification: A Machine Learning Approach

Loris Di Natale, Muhammad Zakwan, Bratislav Svetozarevic, Philipp Heer, Giancarlo Ferrari-Trecate, Colin N. Jones

Machine Learning (ML) and linear System Identification (SI) have been historically developed independently. In this paper, we leverage well-established ML tools - especially the automatic differentiation framework - to introduce SIMBa, a family of discrete linear multi-step-ahead state-space SI methods using backpropagation. SIMBa relies on a novel Linear-Matrix-Inequality-based free parametrization of Schur matrices to ensure the stability of the identified model. We show how SIMBa generally outperforms traditional linear state-space SI methods, and sometimes significantly, although at the price of a higher computational burden. This performance gap is particularly remarkable compared to other SI methods with stability guarantees, where the gain is frequently above 25% in our investigations, hinting at SIMBa's ability to simultaneously achieve state-of-the-art fitting performance and enforce stability. Interestingly, these observations hold for a wide variety of input-output systems and on both simulated and real-world data, showcasing the flexibility of the proposed approach. We postulate that this new SI paradigm presents a great extension potential to identify structured nonlinear models from data, and we hence open-source SIMBa on https://github.com/Cemempamoi/simba.

comment: Accepted at ECC 2024

♻ ☆ DeepMachining: Online Prediction of Machining Errors of Lathe Machines

Xiang-Li Lu, Hwai-Jung Hsu, Che-Wei Chou, H. T. Kung, Chen-Hsin Lee

We describe DeepMachining, a deep learning-based AI system for online prediction of machining errors of lathe machine operations. We have built and evaluated DeepMachining based on manufacturing data from factories. Specifically, we first pretrain a deep learning model for a given lathe machine's operations to learn the salient features of machining states. Then, we fine-tune the pretrained model to adapt to specific machining tasks. We demonstrate that DeepMachining achieves high prediction accuracy for multiple tasks that involve different workpieces and cutting tools. To the best of our knowledge, this work is one of the first factory experiments using pre-trained deep-learning models to predict machining errors of lathe machines.

♻ ☆ RetroBridge: Modeling Retrosynthesis with Markov Bridges

Ilia Igashov, Arne Schneuing, Marwin Segler, Michael Bronstein, Bruno Correia

Retrosynthesis planning is a fundamental challenge in chemistry which aims at designing reaction pathways from commercially available starting materials to a target molecule. Each step in multi-step retrosynthesis planning requires accurate prediction of possible precursor molecules given the target molecule and confidence estimates to guide heuristic search algorithms. We model single-step retrosynthesis planning as a distribution learning problem in a discrete state space. First, we introduce the Markov Bridge Model, a generative framework aimed to approximate the dependency between two intractable discrete distributions accessible via a finite sample of coupled data points. Our framework is based on the concept of a Markov bridge, a Markov process pinned at its endpoints. Unlike diffusion-based methods, our Markov Bridge Model does not need a tractable noise distribution as a sampling proxy and directly operates on the input product molecules as samples from the intractable prior distribution. We then address the retrosynthesis planning problem with our novel framework and introduce RetroBridge, a template-free retrosynthesis modeling approach that achieves state-of-the-art results on standard evaluation benchmarks.

♻ ☆ ChIRAAG: ChatGPT Informed Rapid and Automated Assertion Generation

Bhabesh Mali, Karthik Maddala, Sweeya Reddy, Vatsal Gupta, Chandan Karfa, Ramesh Karri

System Verilog Assertion (SVA) formulation- a critical yet complex task is a prerequisite in the Formal Property Verification (FPV) process. Traditionally, SVA formulation involves expert-driven interpretation of specifications, which is timeconsuming and prone to human error. However, LLM-informed automatic assertion generation is gaining interest. We designeda novel framework called ChIRAAG, based on OpenAI GPT4, to generate SVA assertions from natural language specifications. ChIRAAG constitutes the systematic breakdown of design specifications into a standardized format, further generating assertions from formatted specifications using LLM. Furthermore, we developed testbenches to verify/validate the LLM-generated assertions. Automatic feedback of log files from the simulation tool to the LLM ensures that the framework can generate correc SVAs automatically. Only 33% of LLM-generated raw assertions had errors. Our results on OpenTitan designs shows that LLMs can streamline and assist engineers in the assertion generation process, reshaping verification workflows.

comment: 6 pages, 5 figures and 2 table

♻ ☆ Towards Low-Energy Adaptive Personalization for Resource-Constrained Devices

Yushan Huang, Josh Millar, Yuxuan Long, Yuchen Zhao, Hamed Hadaddi

The personalization of machine learning (ML) models to address data drift is a significant challenge in the context of Internet of Things (IoT) applications. Presently, most approaches focus on fine-tuning either the full base model or its last few layers to adapt to new data, while often neglecting energy costs. However, various types of data drift exist, and fine-tuning the full base model or the last few layers may not result in optimal performance in certain scenarios. We propose Target Block Fine-Tuning (TBFT), a low-energy adaptive personalization framework designed for resource-constrained devices. We categorize data drift and personalization into three types: input-level, feature-level, and output-level. For each type, we fine-tune different blocks of the model to achieve optimal performance with reduced energy costs. Specifically, input-, feature-, and output-level correspond to fine-tuning the front, middle, and rear blocks of the model. We evaluate TBFT on a ResNet model, three datasets, three different training sizes, and a Raspberry Pi. Compared with the $Block Avg$, where each block is fine-tuned individually and their performance improvements are averaged, TBFT exhibits an improvement in model accuracy by an average of 15.30% whilst saving 41.57% energy consumption on average compared with full fine-tuning.

comment: Accepetd to The 4th Workshop on Machine Learning and Systems (EuroMLSys '24)

♻ ☆ FedCSD: A Federated Learning Based Approach for Code-Smell Detection

Sadi Alawadi, Khalid Alkharabsheh, Fahed Alkhabbas, Victor Kebande, Feras M. Awaysheh, Fabio Palomba, Mohammed Awad

This paper proposes a Federated Learning Code Smell Detection (FedCSD) approach that allows organizations to collaboratively train federated ML models while preserving their data privacy. These assertions have been supported by three experiments that have significantly leveraged three manually validated datasets aimed at detecting and examining different code smell scenarios. In experiment 1, which was concerned with a centralized training experiment, dataset two achieved the lowest accuracy (92.30%) with fewer smells, while datasets one and three achieved the highest accuracy with a slight difference (98.90% and 99.5%, respectively). This was followed by experiment 2, which was concerned with cross-evaluation, where each ML model was trained using one dataset, which was then evaluated over the other two datasets. Results from this experiment show a significant drop in the model's accuracy (lowest accuracy: 63.80\%) where fewer smells exist in the training dataset, which has a noticeable reflection (technical debt) on the model's performance. Finally, the last and third experiments evaluate our approach by splitting the dataset into 10 companies. The ML model was trained on the company's site, then all model-updated weights were transferred to the server. Ultimately, an accuracy of 98.34% was achieved by the global model that has been trained using 10 companies for 100 training rounds. The results reveal a slight difference in the global model's accuracy compared to the highest accuracy of the centralized model, which can be ignored in favour of the global model's comprehensive knowledge, lower training cost, preservation of data privacy, and avoidance of the technical debt problem.

comment: 17 pages, 7 figures, Journal paper

♻ ☆ A Multimodal Approach to Device-Directed Speech Detection with Large Language Models

Dominik Wagner, Alexander Churchill, Siddharth Sigtia, Panayiotis Georgiou, Matt Mirsamadi, Aarshee Mishra, Erik Marchi

Interactions with virtual assistants typically start with a predefined trigger phrase followed by the user command. To make interactions with the assistant more intuitive, we explore whether it is feasible to drop the requirement that users must begin each command with a trigger phrase. We explore this task in three ways: First, we train classifiers using only acoustic information obtained from the audio waveform. Second, we take the decoder outputs of an automatic speech recognition (ASR) system, such as 1-best hypotheses, as input features to a large language model (LLM). Finally, we explore a multimodal system that combines acoustic and lexical features, as well as ASR decoder signals in an LLM. Using multimodal information yields relative equal-error-rate improvements over text-only and audio-only models of up to 39% and 61%. Increasing the size of the LLM and training with low-rank adaption leads to further relative EER reductions of up to 18% on our dataset.

comment: arXiv admin note: text overlap with arXiv:2312.03632

♻ ☆ FedCau: A Proactive Stop Policy for Communication and Computation Efficient Federated Learning

Afsaneh Mahmoudi, Hossein S. Ghadikolaei, José Mairton Barros Da Silva Júnior, Carlo Fischione

This paper investigates efficient distributed training of a Federated Learning~(FL) model over a wireless network of wireless devices. The communication iterations of the distributed training algorithm may be substantially deteriorated or even blocked by the effects of the devices' background traffic, packet losses, congestion, or latency. We abstract the communication-computation impacts as an `iteration cost' and propose a cost-aware causal FL algorithm~(FedCau) to tackle this problem. We propose an iteration-termination method that trade-offs the training performance and networking costs. We apply our approach when clients use the slotted-ALOHA, the carrier-sense multiple access with collision avoidance~(CSMA/CA), and the orthogonal frequency-division multiple access~(OFDMA) protocols. We show that, given a total cost budget, the training performance degrades as either the background communication traffic or the dimension of the training problem increases. Our results demonstrate the importance of proactively designing optimal cost-efficient stopping criteria to avoid unnecessary communication-computation costs to achieve only a marginal FL training improvement. We validate our method by training and testing FL over the MNIST dataset. Finally, we apply our approach to existing communication efficient FL methods from the literature, achieving further efficiency. We conclude that cost-efficient stopping criteria are essential for the success of practical FL over wireless networks.

♻ ☆ DreamComposer: Controllable 3D Object Generation via Multi-View Conditions

Yunhan Yang, Yukun Huang, Xiaoyang Wu, Yuan-Chen Guo, Song-Hai Zhang, Hengshuang Zhao, Tong He, Xihui Liu

Utilizing pre-trained 2D large-scale generative models, recent works are capable of generating high-quality novel views from a single in-the-wild image. However, due to the lack of information from multiple views, these works encounter difficulties in generating controllable novel views. In this paper, we present DreamComposer, a flexible and scalable framework that can enhance existing view-aware diffusion models by injecting multi-view conditions. Specifically, DreamComposer first uses a view-aware 3D lifting module to obtain 3D representations of an object from multiple views. Then, it renders the latent features of the target view from 3D representations with the multi-view feature fusion module. Finally the target view features extracted from multi-view inputs are injected into a pre-trained diffusion model. Experiments show that DreamComposer is compatible with state-of-the-art diffusion models for zero-shot novel view synthesis, further enhancing them to generate high-fidelity novel view images with multi-view conditions, ready for controllable 3D object reconstruction and various other applications.

comment: Project Page: https://yhyang-myron.github.io/DreamComposer/

♻ ☆ Transport meets Variational Inference: Controlled Monte Carlo Diffusions ICML

Francisco Vargas, Shreyas Padhy, Denis Blessing, Nikolas Nüsken

Connecting optimal transport and variational inference, we present a principled and systematic framework for sampling and generative modelling centred around divergences on path space. Our work culminates in the development of the \emph{Controlled Monte Carlo Diffusion} sampler (CMCD) for Bayesian computation, a score-based annealing technique that crucially adapts both forward and backward dynamics in a diffusion model. On the way, we clarify the relationship between the EM-algorithm and iterative proportional fitting (IPF) for Schr{\"o}dinger bridges, deriving as well a regularised objective that bypasses the iterative bottleneck of standard IPF-updates. Finally, we show that CMCD has a strong foundation in the Jarzinsky and Crooks identities from statistical physics, and that it convincingly outperforms competing approaches across a wide array of experiments.

comment: Workshop on New Frontiers in Learning, Control, and Dynamical Systems at the International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA, 2023

♻ ☆ P2ANet: A Dataset and Benchmark for Dense Action Detection from Table Tennis Match Broadcasting Videos

Jiang Bian, Xuhong Li, Tao Wang, Qingzhong Wang, Jun Huang, Chen Liu, Jun Zhao, Feixiang Lu, Dejing Dou, Haoyi Xiong

While deep learning has been widely used for video analytics, such as video classification and action detection, dense action detection with fast-moving subjects from sports videos is still challenging. In this work, we release yet another sports video benchmark \TheName{} for \emph{\underline{P}}ing \emph{\underline{P}}ong-\emph{\underline{A}}ction detection, which consists of 2,721 video clips collected from the broadcasting videos of professional table tennis matches in World Table Tennis Championships and Olympiads. We work with a crew of table tennis professionals and referees on a specially designed annotation toolbox to obtain fine-grained action labels (in 14 classes) for every ping-pong action that appeared in the dataset, and formulate two sets of action detection problems -- \emph{action localization} and \emph{action recognition}. We evaluate a number of commonly-seen action recognition (e.g., TSM, TSN, Video SwinTransformer, and Slowfast) and action localization models (e.g., BSN, BSN++, BMN, TCANet), using \TheName{} for both problems, under various settings. These models can only achieve 48\% area under the AR-AN curve for localization and 82\% top-one accuracy for recognition since the ping-pong actions are dense with fast-moving subjects but broadcasting videos are with only 25 FPS. The results confirm that \TheName{} is still a challenging task and can be used as a special benchmark for dense action detection from videos.

♻ ☆ An Implicit GNN Solver for Poisson-like problems

Matthieu Nastorg, Michele Alessandro Bucci, Thibault Faney, Jean-Marc Gratien, Guillaume Charpiat, Marc Schoenauer

This paper presents $\Psi$-GNN, a novel Graph Neural Network (GNN) approach for solving the ubiquitous Poisson PDE problems with mixed boundary conditions. By leveraging the Implicit Layer Theory, $\Psi$-GNN models an "infinitely" deep network, thus avoiding the empirical tuning of the number of required Message Passing layers to attain the solution. Its original architecture explicitly takes into account the boundary conditions, a critical prerequisite for physical applications, and is able to adapt to any initially provided solution. $\Psi$-GNN is trained using a "physics-informed" loss, and the training process is stable by design, and insensitive to its initialization. Furthermore, the consistency of the approach is theoretically proven, and its flexibility and generalization efficiency are experimentally demonstrated: the same learned model can accurately handle unstructured meshes of various sizes, as well as different boundary conditions. To the best of our knowledge, $\Psi$-GNN is the first physics-informed GNN-based method that can handle various unstructured domains, boundary conditions and initial solutions while also providing convergence guarantees.

♻ ☆ Ensemble learning for Physics Informed Neural Networks: a Gradient Boosting approach

Zhiwei Fang, Sifan Wang, Paris Perdikaris

While the popularity of physics-informed neural networks (PINNs) is steadily rising, to this date, PINNs have not been successful in simulating multi-scale and singular perturbation problems. In this work, we present a new training paradigm referred to as "gradient boosting" (GB), which significantly enhances the performance of physics informed neural networks (PINNs). Rather than learning the solution of a given PDE using a single neural network directly, our algorithm employs a sequence of neural networks to achieve a superior outcome. This approach allows us to solve problems presenting great challenges for traditional PINNs. Our numerical experiments demonstrate the effectiveness of our algorithm through various benchmarks, including comparisons with finite element methods and PINNs. Furthermore, this work also unlocks the door to employing ensemble learning techniques in PINNs, providing opportunities for further improvement in solving PDEs.

♻ ☆ Bayesian data-driven discovery of partial differential equations with variable coefficients

Aoxue Chen, Yifan Du, Liyao Mars Gao, Guang Lin

The discovery of Partial Differential Equations (PDEs) is an essential task for applied science and engineering. However, data-driven discovery of PDEs is generally challenging, primarily stemming from the sensitivity of the discovered equation to noise and the complexities of model selection. In this work, we propose an advanced Bayesian sparse learning algorithm for PDE discovery with variable coefficients, predominantly when the coefficients are spatially or temporally dependent. Specifically, we apply threshold Bayesian group Lasso regression with a spike-and-slab prior (tBGL-SS) and leverage a Gibbs sampler for Bayesian posterior estimation of PDE coefficients. This approach not only enhances the robustness of point estimation with valid uncertainty quantification but also relaxes the computational burden from Bayesian inference through the integration of coefficient thresholds as an approximate MCMC method. Moreover, from the quantified uncertainties, we propose a Bayesian total error bar criteria for model selection, which outperforms classic metrics including the root mean square and the Akaike information criterion. The capability of this method is illustrated by the discovery of several classical benchmark PDEs with spatially or temporally varying coefficients from solution data obtained from the reference simulations. In the experiments, we show that the tBGL-SS method is more robust than the baseline methods under noisy environments and provides better model selection criteria along the regularization path.

♻ ☆ Graph Signal Diffusion Model for Collaborative Filtering SIGIR 2024

Yunqin Zhu, Chao Wang, Qi Zhang, Hui Xiong

Collaborative filtering is a critical technique in recommender systems. Among various methods, an increasingly popular paradigm is to reconstruct user-item interactions based on the historical observations. This can be viewed as a conditional generative task, where recently developed diffusion model demonstrates great potential. However, existing studies on diffusion models lack effective solutions for modeling implicit feedback data. Particularly, the isotropic nature of the standard diffusion process fails to account for the heterogeneous dependencies among items, leading to a misalignment with the graphical structure of the interaction space. Meanwhile, random noise destroying personalized information in interaction vectors, causing difficulty in reverse reconstruction. In this paper, we make novel adaptions of diffusion model and propose Graph Signal Diffusion Model for Collaborative Filtering (named GiffCF). To better represent the high-dimensional and sparse distribution of implicit feedback, we define a generalized form of denoising diffusion using heat equation on the item-item similarity graph. Our forward process smooths interaction signals with an advanced family of graph filters. Hence, instead of losing information, it involves item-item similarities as beneficial prior knowledge for recommendation. To reconstruct high-quality interactions, our reverse process iteratively refines and sharpens preference signals in a deterministic manner, where the update direction is conditioned on the user history and computed from a carefully designed two-stage denoiser. Finally, through extensive experiments, we show that GiffCF effectively leverages the advantages of both diffusion model and graph signal processing, and achieves state-of-the-art performance on three benchmark datasets.

comment: 11 pages, 8 figures, Accepted by SIGIR 2024

♻ ☆ Riemannian Laplace Approximation with the Fisher Metric AISTATS 2024

Hanlin Yu, Marcelo Hartmann, Bernardo Williams, Mark Girolami, Arto Klami

Laplace's method approximates a target density with a Gaussian distribution at its mode. It is computationally efficient and asymptotically exact for Bayesian inference due to the Bernstein-von Mises theorem, but for complex targets and finite-data posteriors it is often too crude an approximation. A recent generalization of the Laplace Approximation transforms the Gaussian approximation according to a chosen Riemannian geometry providing a richer approximation family, while still retaining computational efficiency. However, as shown here, its properties depend heavily on the chosen metric, indeed the metric adopted in previous work results in approximations that are overly narrow as well as being biased even at the limit of infinite data. We correct this shortcoming by developing the approximation family further, deriving two alternative variants that are exact at the limit of infinite data, extending the theoretical analysis of the method, and demonstrating practical improvements in a range of experiments.

comment: AISTATS 2024, with additional fixes

♻ ☆ DISL: Fueling Research with A Large Dataset of Solidity Smart Contracts

Gabriele Morello, Mojtaba Eshghie, Sofia Bobadilla, Martin Monperrus

The DISL dataset features a collection of $514,506$ unique Solidity files that have been deployed to Ethereum mainnet. It caters to the need for a large and diverse dataset of real-world smart contracts. DISL serves as a resource for developing machine learning systems and for benchmarking software engineering tools designed for smart contracts. By aggregating every verified smart contract from Etherscan up to January 15, 2024, DISL surpasses existing datasets in size and recency.

♻ ☆ Masked Autoencoders Are Robust Neural Architecture Search Learners

Yiming Hu, Xiangxiang Chu, Bo Zhang

Neural Architecture Search (NAS) currently relies heavily on labeled data, which is both expensive and time-consuming to acquire. In this paper, we propose a novel NAS framework based on Masked Autoencoders (MAE) that eliminates the need for labeled data during the search process. By replacing the supervised learning objective with an image reconstruction task, our approach enables the robust discovery of network architectures without compromising performance and generalization ability. Additionally, we address the problem of performance collapse encountered in the widely-used Differentiable Architecture Search (DARTS) method in the unsupervised paradigm by introducing a multi-scale decoder. Through extensive experiments conducted on various search spaces and datasets, we demonstrate the effectiveness and robustness of the proposed method, providing empirical evidence of its superiority over baseline approaches.

♻ ☆ Domain-Aware Fine-Tuning: Enhancing Neural Network Adaptability

Seokhyeon Ha, Sunbeom Jung, Jungwoo Lee

Fine-tuning pre-trained neural network models has become a widely adopted approach across various domains. However, it can lead to the distortion of pre-trained feature extractors that already possess strong generalization capabilities. Mitigating feature distortion during adaptation to new target domains is crucial. Recent studies have shown promising results in handling feature distortion by aligning the head layer on in-distribution datasets before performing fine-tuning. Nonetheless, a significant limitation arises from the treatment of batch normalization layers during fine-tuning, leading to suboptimal performance. In this paper, we propose Domain-Aware Fine-Tuning (DAFT), a novel approach that incorporates batch normalization conversion and the integration of linear probing and fine-tuning. Our batch normalization conversion method effectively mitigates feature distortion by reducing modifications to the neural network during fine-tuning. Additionally, we introduce the integration of linear probing and fine-tuning to optimize the head layer with gradual adaptation of the feature extractor. By leveraging batch normalization layers and integrating linear probing and fine-tuning, our DAFT significantly mitigates feature distortion and achieves improved model performance on both in-distribution and out-of-distribution datasets. Extensive experiments demonstrate that our method outperforms other baseline methods, demonstrating its effectiveness in not only improving performance but also mitigating feature distortion.

♻ ☆ Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models

Joao Carvalho, An T. Le, Mark Baierl, Dorothea Koert, Jan Peters

Learning priors on trajectory distributions can help accelerate robot motion planning optimization. Given previously successful plans, learning trajectory generative models as priors for a new planning problem is highly desirable. Prior works propose several ways on utilizing this prior to bootstrapping the motion planning problem. Either sampling the prior for initializations or using the prior distribution in a maximum-a-posterior formulation for trajectory optimization. In this work, we propose learning diffusion models as priors. We then can sample directly from the posterior trajectory distribution conditioned on task goals, by leveraging the inverse denoising process of diffusion models. Furthermore, diffusion has been recently shown to effectively encode data multimodality in high-dimensional settings, which is particularly well-suited for large trajectory dataset. To demonstrate our method efficacy, we compare our proposed method - Motion Planning Diffusion - against several baselines in simulated planar robot and 7-dof robot arm manipulator environments. To assess the generalization capabilities of our method, we test it in environments with previously unseen obstacles. Our experiments show that diffusion models are strong priors to encode high-dimensional trajectory distributions of robot motions.

♻ ☆ A Lightweight and Gradient-Stable Neural Layer

Yueyao Yu, Yin Zhang

To enhance resource efficiency and model deployability of neural networks, we propose a neural-layer architecture based on Householder weighting and absolute-value activating, called Householder-absolute neural layer or simply Han-layer. Compared to a fully connected layer with $d$-neurons and $d$ outputs, a Han-layer reduces the number of parameters and the corresponding computational complexity from $O(d^2)$ to $O(d)$. {The Han-layer structure guarantees that the Jacobian of the layer function is always orthogonal, thus ensuring gradient stability (i.e., free of gradient vanishing or exploding issues) for any Han-layer sub-networks.} Extensive numerical experiments show that one can strategically use Han-layers to replace fully connected (FC) layers, reducing the number of model parameters while maintaining or even improving the generalization performance. We will also showcase the capabilities of the Han-layer architecture on a few small stylized models, and discuss its current limitations.

♻ ☆ Brain Networks and Intelligence: A Graph Neural Network Based Approach to Resting State fMRI Data

Bishal Thapaliya, Esra Akbas, Jiayu Chen, Raam Sapkota, Bhaskar Ray, Pranav Suresh, Vince Calhoun, Jingyu Liu

Resting-state functional magnetic resonance imaging (rsfMRI) is a powerful tool for investigating the relationship between brain function and cognitive processes as it allows for the functional organization of the brain to be captured without relying on a specific task or stimuli. In this paper, we present a novel modeling architecture called BrainRGIN for predicting intelligence (fluid, crystallized, and total intelligence) using graph neural networks on rsfMRI derived static functional network connectivity matrices. Extending from the existing graph convolution networks, our approach incorporates a clustering-based embedding and graph isomorphism network in the graph convolutional layer to reflect the nature of the brain sub-network organization and efficient network expression, in combination with TopK pooling and attention-based readout functions. We evaluated our proposed architecture on a large dataset, specifically the Adolescent Brain Cognitive Development Dataset, and demonstrated its effectiveness in predicting individual differences in intelligence. Our model achieved lower mean squared errors and higher correlation scores than existing relevant graph architectures and other traditional machine learning models for all of the intelligence prediction tasks. The middle frontal gyrus exhibited a significant contribution to both fluid and crystallized intelligence, suggesting their pivotal role in these cognitive processes. Total composite scores identified a diverse set of brain regions to be relevant which underscores the complex nature of total intelligence.

♻ ☆ All Rivers Run to the Sea: Private Learning with Asymmetric Flows CVPR 2024

Yue Niu, Ramy E. Ali, Saurav Prakash, Salman Avestimehr

Data privacy is of great concern in cloud machine-learning service platforms, when sensitive data are exposed to service providers. While private computing environments (e.g., secure enclaves), and cryptographic approaches (e.g., homomorphic encryption) provide strong privacy protection, their computing performance still falls short compared to cloud GPUs. To achieve privacy protection with high computing performance, we propose Delta, a new private training and inference framework, with comparable model performance as non-private centralized training. Delta features two asymmetric data flows: the main information-sensitive flow and the residual flow. The main part flows into a small model while the residuals are offloaded to a large model. Specifically, Delta embeds the information-sensitive representations into a low-dimensional space while pushing the information-insensitive part into high-dimension residuals. To ensure privacy protection, the low-dimensional information-sensitive part is secured and fed to a small model in a private environment. On the other hand, the residual part is sent to fast cloud GPUs, and processed by a large model. To further enhance privacy and reduce the communication cost, Delta applies a random binary quantization technique along with a DP-based technique to the residuals before sharing them with the public platform. We theoretically show that Delta guarantees differential privacy in the public environment and greatly reduces the complexity in the private environment. We conduct empirical analyses on CIFAR-10, CIFAR-100 and ImageNet datasets and ResNet-18 and ResNet-34, showing that Delta achieves strong privacy protection, fast training, and inference without significantly compromising the model utility.

comment: Camera-ready for CVPR 2024

♻ ☆ NaNa and MiGu: Semantic Data Augmentation Techniques to Enhance Protein Classification in Graph Neural Networks

Yi-Shan Lan, Pin-Yu Chen, Tsung-Yi Ho

Protein classification tasks are essential in drug discovery. Real-world protein structures are dynamic, which will determine the properties of proteins. However, the existing machine learning methods, like ProNet (Wang et al., 2022a), only access limited conformational characteristics and protein side-chain features, leading to impractical protein structure and inaccuracy of protein classes in their predictions. In this paper, we propose novel semantic data augmentation methods, Novel Augmentation of New Node Attributes (NaNa), and Molecular Interactions and Geometric Upgrading (MiGu) to incorporate backbone chemical and side-chain biophysical information into protein classification tasks and a co-embedding residual learning framework. Specifically, we leverage molecular biophysical, secondary structure, chemical bonds, and ionic features of proteins to facilitate protein classification tasks. Furthermore, our semantic augmentation methods and the co-embedding residual learning framework can improve the performance of GIN (Xu et al., 2019) on EC and Fold datasets (Bairoch, 2000; Andreeva et al., 2007) by 16.41% and 11.33% respectively. Our code is available at https://github.com/r08b46009/Code_for_MIGU_NANA/tree/main.

♻ ☆ Graph Generation with $K^2$-trees ICLR

Yunhui Jang, Dongwoo Kim, Sungsoo Ahn

Generating graphs from a target distribution is a significant challenge across many domains, including drug discovery and social network analysis. In this work, we introduce a novel graph generation method leveraging $K^2$-tree representation, originally designed for lossless graph compression. The $K^2$-tree representation {encompasses inherent hierarchy while enabling compact graph generation}. In addition, we make contributions by (1) presenting a sequential $K^2$-treerepresentation that incorporates pruning, flattening, and tokenization processes and (2) introducing a Transformer-based architecture designed to generate the sequence by incorporating a specialized tree positional encoding scheme. Finally, we extensively evaluate our algorithm on four general and two molecular graph datasets to confirm its superiority for graph generation.

comment: International Conference on Learning Representations (ICLR) 2024

♻ ☆ A Simple and Scalable Representation for Graph Generation ICLR

Yunhui Jang, Seul Lee, Sungsoo Ahn

Recently, there has been a surge of interest in employing neural networks for graph generation, a fundamental statistical learning problem with critical applications like molecule design and community analysis. However, most approaches encounter significant limitations when generating large-scale graphs. This is due to their requirement to output the full adjacency matrices whose size grows quadratically with the number of nodes. In response to this challenge, we introduce a new, simple, and scalable graph representation named gap encoded edge list (GEEL) that has a small representation size that aligns with the number of edges. In addition, GEEL significantly reduces the vocabulary size by incorporating the gap encoding and bandwidth restriction schemes. GEEL can be autoregressively generated with the incorporation of node positional encoding, and we further extend GEEL to deal with attributed graphs by designing a new grammar. Our findings reveal that the adoption of this compact representation not only enhances scalability but also bolsters performance by simplifying the graph generation process. We conduct a comprehensive evaluation across ten non-attributed and two molecular graph generation tasks, demonstrating the effectiveness of GEEL.

comment: International Conference on Learning Representations (ICLR) 2024

♻ ☆ Identification of Craving Maps among Marijuana Users via the Analysis of Functional Brain Networks with High-Order Attention Graph Neural Networks

Jun-En Ding, Shihao Yang, Anna Zilverstand, Feng Liu

The excessive consumption of marijuana can induce substantial psychological and social consequences. In this investigation, we propose an elucidative framework termed high-order graph attention neural networks (HOGANN) for the classification of Marijuana addiction, coupled with an analysis of localized brain network communities exhibiting abnormal activities among chronic marijuana users. HOGANN integrates dynamic intrinsic functional brain networks, estimated from resting-state functional magnetic resonance imaging (rs-fMRI), using long short-term memory (LSTM) to capture temporal network dynamics. We employ a high-order attention module for information fusion and message passing among neighboring nodes, enhancing the network community analysis. Our model is validated across two distinct data cohorts, yielding substantially higher classification accuracy than benchmark algorithms. Furthermore, we discern the most pertinent subnetworks and cognitive regions affected by persistent marijuana consumption, indicating adverse effects on functional brain networks, particularly within the dorsal attention and frontoparietal networks. Intriguingly, our model demonstrates superior performance in cohorts exhibiting prolonged dependence, implying that prolonged marijuana usage induces more pronounced alterations in brain networks. The model proficiently identifies craving brain maps, thereby delineating critical brain regions for analysis.

♻ ☆ Troika: Multi-Path Cross-Modal Traction for Compositional Zero-Shot Learning CVPR 2024

Siteng Huang, Biao Gong, Yutong Feng, Min Zhang, Yiliang Lv, Donglin Wang

Recent compositional zero-shot learning (CZSL) methods adapt pre-trained vision-language models (VLMs) by constructing trainable prompts only for composed state-object pairs. Relying on learning the joint representation of seen compositions, these methods ignore the explicit modeling of the state and object, thus limiting the exploitation of pre-trained knowledge and generalization to unseen compositions. With a particular focus on the universality of the solution, in this work, we propose a novel paradigm for CZSL models that establishes three identification branches (i.e., Multi-Path) to jointly model the state, object, and composition. The presented Troika is our implementation that aligns the branch-specific prompt representations with decomposed visual features. To calibrate the bias between semantically similar multi-modal representations, we further devise a Cross-Modal Traction module into Troika that shifts the prompt representation towards the current visual content. We conduct extensive experiments on three popular benchmarks, where our method significantly outperforms existing methods in both closed-world and open-world settings. The code will be available at https://github.com/bighuang624/Troika.

comment: CVPR 2024

♻ ☆ Prediction Error Estimation in Random Forests

Ian Krupkin, Johanna Hardin

In this paper, error estimates of classification Random Forests are quantitatively assessed. Based on the initial theoretical framework built by Bates et al. (2023), the true error rate and expected error rate are theoretically and empirically investigated in the context of a variety of error estimation methods common to Random Forests. We show that in the classification case, Random Forests' estimates of prediction error is closer on average to the true error rate instead of the average prediction error. This is opposite the findings of Bates et al. (2023) which are given for logistic regression. We further show that our result holds across different error estimation strategies such as cross-validation, bagging, and data splitting.

comment: arXiv admin note: text overlap with arXiv:2104.00673 by other authors

♻ ☆ Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators

Yinhong Liu, Han Zhou, Zhijiang Guo, Ehsan Shareghi, Ivan Vulić, Anna Korhonen, Nigel Collier

Large Language Models (LLMs) have demonstrated promising capabilities as automatic evaluators in assessing the quality of generated natural language. However, LLMs still exhibit biases in evaluation and often struggle to generate coherent evaluations that align with human assessments. In this work, we first conduct a systematic study of the misalignment between LLM evaluators and human judgement, revealing that existing calibration methods aimed at mitigating biases are insufficient for effectively aligning LLM evaluators. Inspired by the use of preference data in RLHF, we formulate the evaluation as a ranking problem and introduce Pairwise-preference Search (PairS), an uncertainty-guided search method that employs LLMs to conduct pairwise comparisons and efficiently ranks candidate texts. PairS achieves state-of-the-art performance on representative evaluation tasks and demonstrates significant improvements over direct scoring. Furthermore, we provide insights into the role of pairwise preference in quantifying the transitivity of LLMs and demonstrate how PairS benefits from calibration.

♻ ☆ Learning Flexible Body Collision Dynamics with Hierarchical Contact Mesh Transformer ICLR 2024

Youn-Yeol Yu, Jeongwhan Choi, Woojin Cho, Kookjin Lee, Nayong Kim, Kiseok Chang, Chang-Seung Woo, Ilho Kim, Seok-Woo Lee, Joon-Young Yang, Sooyoung Yoon, Noseong Park

Recently, many mesh-based graph neural network (GNN) models have been proposed for modeling complex high-dimensional physical systems. Remarkable achievements have been made in significantly reducing the solving time compared to traditional numerical solvers. These methods are typically designed to i) reduce the computational cost in solving physical dynamics and/or ii) propose techniques to enhance the solution accuracy in fluid and rigid body dynamics. However, it remains under-explored whether they are effective in addressing the challenges of flexible body dynamics, where instantaneous collisions occur within a very short timeframe. In this paper, we present Hierarchical Contact Mesh Transformer (HCMT), which uses hierarchical mesh structures and can learn long-range dependencies (occurred by collisions) among spatially distant positions of a body -- two close positions in a higher-level mesh correspond to two distant positions in a lower-level mesh. HCMT enables long-range interactions, and the hierarchical mesh structure quickly propagates collision effects to faraway positions. To this end, it consists of a contact mesh Transformer and a hierarchical mesh Transformer (CMT and HMT, respectively). Lastly, we propose a flexible body dynamics dataset, consisting of trajectories that reflect experimental settings frequently used in the display industry for product designs. We also compare the performance of several baselines using well-known benchmark datasets. Our results show that HCMT provides significant performance improvements over existing methods. Our code is available at https://github.com/yuyudeep/hcmt.

comment: Accepted at ICLR 2024

♻ ☆ PPI++: Efficient Prediction-Powered Inference

Anastasios N. Angelopoulos, John C. Duchi, Tijana Zrnic

We present PPI++: a computationally lightweight methodology for estimation and inference based on a small labeled dataset and a typically much larger dataset of machine-learning predictions. The methods automatically adapt to the quality of available predictions, yielding easy-to-compute confidence sets -- for parameters of any dimensionality -- that always improve on classical intervals using only the labeled data. PPI++ builds on prediction-powered inference (PPI), which targets the same problem setting, improving its computational and statistical efficiency. Real and synthetic experiments demonstrate the benefits of the proposed adaptations.

comment: Code available at https://github.com/aangelopoulos/ppi_py

♻ ☆ Improving Opioid Use Disorder Risk Modelling through Behavioral and Genetic Feature Integration

Sybille Légitime, Kaustubh Prabhu, Devin McConnell, Bing Wang, Dipak K. Dey, Derek Aguiar

Opioids are an effective analgesic for acute and chronic pain, but also carry a considerable risk of addiction leading to millions of opioid use disorder (OUD) cases and tens of thousands of premature deaths in the United States yearly. Estimating OUD risk prior to prescription could improve the efficacy of treatment regimens, monitoring programs, and intervention strategies, but risk estimation is typically based on self-reported data or questionnaires. We develop an experimental design and computational methods that combine genetic variants associated with OUD with behavioral features extracted from GPS and Wi-Fi spatiotemporal coordinates to assess OUD risk. Since both OUD mobility and genetic data do not exist for the same cohort, we develop algorithms to (1) generate mobility features from empirical distributions and (2) synthesize mobility and genetic samples assuming an expected level of disease co-occurrence. We show that integrating genetic and mobility modalities improves risk modelling using classification accuracy, area under the precision-recall and receiver operator characteristic curves, and $F_1$ score. Interpreting the fitted models suggests that mobility features have more influence on OUD risk, although the genetic contribution was significant, particularly in linear models. While there exist concerns with respect to privacy, security, bias, and generalizability that must be evaluated in clinical trials before being implemented in practice, our framework provides preliminary evidence that behavioral and genetic features may improve OUD risk estimation to assist with personalized clinical decision-making.

comment: 32 pages (including References section), 8 figures. Under review by PLOS One

♻ ☆ Opinion Market Model: Stemming Far-Right Opinion Spread using Positive Interventions AAAI

Pio Calderon, Rohit Ram, Marian-Andrei Rizoiu

Online extremism has severe societal consequences, including normalizing hate speech, user radicalization, and increased social divisions. Various mitigation strategies have been explored to address these consequences. One such strategy uses positive interventions: controlled signals that add attention to the opinion ecosystem to boost certain opinions. To evaluate the effectiveness of positive interventions, we introduce the Opinion Market Model (OMM), a two-tier online opinion ecosystem model that considers both inter-opinion interactions and the role of positive interventions. The size of the opinion attention market is modeled in the first tier using the multivariate discrete-time Hawkes process; in the second tier, opinions cooperate and compete for market share, given limited attention using the market share attraction model. We demonstrate the convergence of our proposed estimation scheme on a synthetic dataset. Next, we test OMM on two learning tasks, applying to two real-world datasets to predict attention market shares and uncover latent relationships between online items. The first dataset comprises Facebook and Twitter discussions containing moderate and far-right opinions about bushfires and climate change. The second dataset captures popular VEVO artists' YouTube and Twitter attention volumes. OMM outperforms the state-of-the-art predictive models on both datasets and captures latent cooperation-competition relations. We uncover (1) self- and cross-reinforcement between far-right and moderate opinions on the bushfires and (2) pairwise artist relations that correlate with real-world interactions such as collaborations and long-lasting feuds. Lastly, we use OMM as a testbed for positive interventions and show how media coverage modulates the spread of far-right opinions.

comment: accepted in the 18th AAAI International Conference on Web and Social Media (ICWSM'24)

♻ ☆ SelfIE: Self-Interpretation of Large Language Model Embeddings

Haozhe Chen, Carl Vondrick, Chengzhi Mao

How do large language models (LLMs) obtain their answers? The ability to explain and control an LLM's reasoning process is key for reliability, transparency, and future model developments. We propose SelfIE (Self-Interpretation of Embeddings), a framework that enables LLMs to interpret their own embeddings in natural language by leveraging their ability to respond to inquiries about a given passage. Capable of interpreting open-world concepts in the hidden embeddings, SelfIE reveals LLM internal reasoning in cases such as making ethical decisions, internalizing prompt injection, and recalling harmful knowledge. SelfIE's text descriptions on hidden embeddings also open up new avenues to control LLM reasoning. We propose Supervised Control, which allows editing open-ended concepts while only requiring gradient computation of individual layer. We extend RLHF to hidden embeddings and propose Reinforcement Control that erases harmful knowledge in LLM without supervision targets.

♻ ☆ Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models ICCV 2023

Songwei Ge, Seungjun Nah, Guilin Liu, Tyler Poon, Andrew Tao, Bryan Catanzaro, David Jacobs, Jia-Bin Huang, Ming-Yu Liu, Yogesh Balaji

Despite tremendous progress in generating high-quality images using diffusion models, synthesizing a sequence of animated frames that are both photorealistic and temporally coherent is still in its infancy. While off-the-shelf billion-scale datasets for image generation are available, collecting similar video data of the same scale is still challenging. Also, training a video diffusion model is computationally much more expensive than its image counterpart. In this work, we explore finetuning a pretrained image diffusion model with video data as a practical solution for the video synthesis task. We find that naively extending the image noise prior to video noise prior in video diffusion leads to sub-optimal performance. Our carefully designed video noise prior leads to substantially better performance. Extensive experimental validation shows that our model, Preserve Your Own Correlation (PYoCo), attains SOTA zero-shot text-to-video results on the UCF-101 and MSR-VTT benchmarks. It also achieves SOTA video generation quality on the small-scale UCF-101 benchmark with a $10\times$ smaller model using significantly less computation than the prior art.

comment: ICCV 2023. Project webpage: https://research.nvidia.com/labs/dir/pyoco

♻ ☆ Near-Optimal Algorithms for Constrained k-Center Clustering with Instance-level Background Knowledge

Longkun Guo, Chaoqi Jia, Kewen Liao, Zhigang Lu, Minhui Xue

Center-based clustering has attracted significant research interest from both theory and practice. In many practical applications, input data often contain background knowledge that can be used to improve clustering results. In this work, we build on widely adopted $k$-center clustering and model its input background knowledge as must-link (ML) and cannot-link (CL) constraint sets. However, most clustering problems including $k$-center are inherently $\mathcal{NP}$-hard, while the more complex constrained variants are known to suffer severer approximation and computation barriers that significantly limit their applicability. By employing a suite of techniques including reverse dominating sets, linear programming (LP) integral polyhedron, and LP duality, we arrive at the first efficient approximation algorithm for constrained $k$-center with the best possible ratio of 2. We also construct competitive baseline algorithms and empirically evaluate our approximation algorithm against them on a variety of real datasets. The results validate our theoretical findings and demonstrate the great advantages of our algorithm in terms of clustering cost, clustering quality, and running time.

♻ ☆ Omega: Optimistic EMA Gradients ICML 2023

Juan Ramirez, Rohan Sukumaran, Quentin Bertrand, Gauthier Gidel

Stochastic min-max optimization has gained interest in the machine learning community with the advancements in GANs and adversarial training. Although game optimization is fairly well understood in the deterministic setting, some issues persist in the stochastic regime. Recent work has shown that stochastic gradient descent-ascent methods such as the optimistic gradient are highly sensitive to noise or can fail to converge. Although alternative strategies exist, they can be prohibitively expensive. We introduce Omega, a method with optimistic-like updates that mitigates the impact of noise by incorporating an EMA of historic gradients in its update rule. We also explore a variation of this algorithm that incorporates momentum. Although we do not provide convergence guarantees, our experiments on stochastic games show that Omega outperforms the optimistic gradient method when applied to linear players.

comment: Oral at the LatinX in AI workshop @ ICML 2023

♻ ☆ Visual Whole-Body Control for Legged Loco-Manipulation

Minghuan Liu, Zixuan Chen, Xuxin Cheng, Yandong Ji, Ruihan Yang, Xiaolong Wang

We study the problem of mobile manipulation using legged robots equipped with an arm, namely legged loco-manipulation. The robot legs, while usually utilized for mobility, offer an opportunity to amplify the manipulation capabilities by conducting whole-body control. That is, the robot can control the legs and the arm at the same time to extend its workspace. We propose a framework that can conduct the whole-body control autonomously with visual observations. Our approach, namely Visual Whole-Body Control(VBC), is composed of a low-level policy using all degrees of freedom to track the end-effector manipulator position and a high-level policy proposing the end-effector position based on visual inputs. We train both levels of policies in simulation and perform Sim2Real transfer for real robot deployment. We perform extensive experiments and show significant improvements over baselines in picking up diverse objects in different configurations (heights, locations, orientations) and environments. Project page: https://wholebody-b1.github.io

comment: The first two authors contribute equally. Project page: https://wholebody-b1.github.io

☆ 2D Gaussian Splatting for Geometrically Accurate Radiance Fields

Binbin Huang, Zehao Yu, Anpei Chen, Andreas Geiger, Shenghua Gao

3D Gaussian Splatting (3DGS) has recently revolutionized radiance field reconstruction, achieving high quality novel view synthesis and fast rendering speed without baking. However, 3DGS fails to accurately represent surfaces due to the multi-view inconsistent nature of 3D Gaussians. We present 2D Gaussian Splatting (2DGS), a novel approach to model and reconstruct geometrically accurate radiance fields from multi-view images. Our key idea is to collapse the 3D volume into a set of 2D oriented planar Gaussian disks. Unlike 3D Gaussians, 2D Gaussians provide view-consistent geometry while modeling surfaces intrinsically. To accurately recover thin surfaces and achieve stable optimization, we introduce a perspective-accurate 2D splatting process utilizing ray-splat intersection and rasterization. Additionally, we incorporate depth distortion and normal consistency terms to further enhance the quality of the reconstructions. We demonstrate that our differentiable renderer allows for noise-free and detailed geometry reconstruction while maintaining competitive appearance quality, fast training speed, and real-time rendering. Our code will be made publicly available.

comment: 12 pages, 12 figures

☆ GTA-HDR: A Large-Scale Synthetic Dataset for HDR Image Reconstruction

Hrishav Bakul Barua, Kalin Stefanov, KokSheik Wong, Abhinav Dhall, Ganesh Krishnasamy

High Dynamic Range (HDR) content (i.e., images and videos) has a broad range of applications. However, capturing HDR content from real-world scenes is expensive and time- consuming. Therefore, the challenging task of reconstructing visually accurate HDR images from their Low Dynamic Range (LDR) counterparts is gaining attention in the vision research community. A major challenge in this research problem is the lack of datasets, which capture diverse scene conditions (e.g., lighting, shadows, weather, locations, landscapes, objects, humans, buildings) and various image features (e.g., color, contrast, saturation, hue, luminance, brightness, radiance). To address this gap, in this paper, we introduce GTA-HDR, a large-scale synthetic dataset of photo-realistic HDR images sampled from the GTA-V video game. We perform thorough evaluation of the proposed dataset, which demonstrates significant qualitative and quantitative improvements of the state-of-the-art HDR image reconstruction methods. Furthermore, we demonstrate the effectiveness of the proposed dataset and its impact on additional computer vision tasks including 3D human pose estimation, human body part segmentation, and holistic scene segmentation. The dataset, data collection pipeline, and evaluation code are available at: https://github.com/HrishavBakulBarua/GTA-HDR.

comment: Submitted to IEEE

☆ DiffH2O: Diffusion-Based Synthesis of Hand-Object Interactions from Textual Descriptions

Sammy Christen, Shreyas Hampali, Fadime Sener, Edoardo Remelli, Tomas Hodan, Eric Sauser, Shugao Ma, Bugra Tekin

Generating natural hand-object interactions in 3D is challenging as the resulting hand and object motions are expected to be physically plausible and semantically meaningful. Furthermore, generalization to unseen objects is hindered by the limited scale of available hand-object interaction datasets. We propose DiffH2O, a novel method to synthesize realistic, one or two-handed object interactions from provided text prompts and geometry of the object. The method introduces three techniques that enable effective learning from limited data. First, we decompose the task into a grasping stage and a text-based interaction stage and use separate diffusion models for each. In the grasping stage, the model only generates hand motions, whereas in the interaction phase both hand and object poses are synthesized. Second, we propose a compact representation that tightly couples hand and object poses. Third, we propose two different guidance schemes to allow more control of the generated motions: grasp guidance and detailed textual guidance. Grasp guidance takes a single target grasping pose and guides the diffusion model to reach this grasp at the end of the grasping stage, which provides control over the grasping pose. Given a grasping motion from this stage, multiple different actions can be prompted in the interaction phase. For textual guidance, we contribute comprehensive text descriptions to the GRAB dataset and show that they enable our method to have more fine-grained control over hand-object interactions. Our quantitative and qualitative evaluation demonstrates that the proposed method outperforms baseline methods and leads to natural hand-object motions. Moreover, we demonstrate the practicality of our framework by utilizing a hand pose estimate from an off-the-shelf pose estimator for guidance, and then sampling multiple different actions in the interaction stage.

comment: Project Page: https://diffh2o.github.io/

☆ GenesisTex: Adapting Image Denoising Diffusion to Texture Space

Chenjian Gao, Boyan Jiang, Xinghui Li, Yingpeng Zhang, Qian Yu

We present GenesisTex, a novel method for synthesizing textures for 3D geometries from text descriptions. GenesisTex adapts the pretrained image diffusion model to texture space by texture space sampling. Specifically, we maintain a latent texture map for each viewpoint, which is updated with predicted noise on the rendering of the corresponding viewpoint. The sampled latent texture maps are then decoded into a final texture map. During the sampling process, we focus on both global and local consistency across multiple viewpoints: global consistency is achieved through the integration of style consistency mechanisms within the noise prediction network, and low-level consistency is achieved by dynamically aligning latent textures. Finally, we apply reference-based inpainting and img2img on denser views for texture refinement. Our approach overcomes the limitations of slow optimization in distillation-based methods and instability in inpainting-based methods. Experiments on meshes from various sources demonstrate that our method surpasses the baseline methods quantitatively and qualitatively.

comment: 12 pages, 10 figures

☆ Makeup Prior Models for 3D Facial Makeup Estimation and Applications CVPR2024

Xingchao Yang, Takafumi Taketomi, Yuki Endo, Yoshihiro Kanamori

In this work, we introduce two types of makeup prior models to extend existing 3D face prior models: PCA-based and StyleGAN2-based priors. The PCA-based prior model is a linear model that is easy to construct and is computationally efficient. However, it retains only low-frequency information. Conversely, the StyleGAN2-based model can represent high-frequency information with relatively higher computational cost than the PCA-based model. Although there is a trade-off between the two models, both are applicable to 3D facial makeup estimation and related applications. By leveraging makeup prior models and designing a makeup consistency module, we effectively address the challenges that previous methods faced in robustly estimating makeup, particularly in the context of handling self-occluded faces. In experiments, we demonstrate that our approach reduces computational costs by several orders of magnitude, achieving speeds up to 180 times faster. In addition, by improving the accuracy of the estimated makeup, we confirm that our methods are highly advantageous for various 3D facial makeup applications such as 3D makeup face reconstruction, user-friendly makeup editing, makeup transfer, and interpolation.

comment: CVPR2024. Project: https://yangxingchao.github.io/makeup-priors-page

☆ AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animation

Huawei Wei, Zejun Yang, Zhisheng Wang

In this study, we propose AniPortrait, a novel framework for generating high-quality animation driven by audio and a reference portrait image. Our methodology is divided into two stages. Initially, we extract 3D intermediate representations from audio and project them into a sequence of 2D facial landmarks. Subsequently, we employ a robust diffusion model, coupled with a motion module, to convert the landmark sequence into photorealistic and temporally consistent portrait animation. Experimental results demonstrate the superiority of AniPortrait in terms of facial naturalness, pose diversity, and visual quality, thereby offering an enhanced perceptual experience. Moreover, our methodology exhibits considerable potential in terms of flexibility and controllability, which can be effectively applied in areas such as facial motion editing or face reenactment. We release code and model weights at https://github.com/scutzzj/AniPortrait

☆ Predicting Perceived Gloss: Do Weak Labels Suffice?

Julia Guerrero-Viu, J. Daniel Subias, Ana Serrano, Katherine R. Storrs, Roland W. Fleming, Belen Masia, Diego Gutierrez

Estimating perceptual attributes of materials directly from images is a challenging task due to their complex, not fully-understood interactions with external factors, such as geometry and lighting. Supervised deep learning models have recently been shown to outperform traditional approaches, but rely on large datasets of human-annotated images for accurate perception predictions. Obtaining reliable annotations is a costly endeavor, aggravated by the limited ability of these models to generalise to different aspects of appearance. In this work, we show how a much smaller set of human annotations ("strong labels") can be effectively augmented with automatically derived "weak labels" in the context of learning a low-dimensional image-computable gloss metric. We evaluate three alternative weak labels for predicting human gloss perception from limited annotated data. Incorporating weak labels enhances our gloss prediction beyond the current state of the art. Moreover, it enables a substantial reduction in human annotation costs without sacrificing accuracy, whether working with rendered images or real photographs.

comment: Computer Graphics Forum (Eurographics 2024)

☆ WordRobe: Text-Guided Generation of Textured 3D Garments

Astitva Srivastava, Pranav Manu, Amit Raj, Varun Jampani, Avinash Sharma

In this paper, we tackle a new and challenging problem of text-driven generation of 3D garments with high-quality textures. We propose "WordRobe", a novel framework for the generation of unposed & textured 3D garment meshes from user-friendly text prompts. We achieve this by first learning a latent representation of 3D garments using a novel coarse-to-fine training strategy and a loss for latent disentanglement, promoting better latent interpolation. Subsequently, we align the garment latent space to the CLIP embedding space in a weakly supervised manner, enabling text-driven 3D garment generation and editing. For appearance modeling, we leverage the zero-shot generation capability of ControlNet to synthesize view-consistent texture maps in a single feed-forward inference step, thereby drastically decreasing the generation time as compared to existing methods. We demonstrate superior performance over current SOTAs for learning 3D garment latent space, garment interpolation, and text-driven texture synthesis, supported by quantitative evaluation and qualitative user study. The unposed 3D garment meshes generated using WordRobe can be directly fed to standard cloth simulation & animation pipelines without any post-processing.

☆ Dr.Hair: Reconstructing Scalp-Connected Hair Strands without Pre-training via Differentiable Rendering of Line Segments CVPR 2024

Yusuke Takimoto, Hikari Takehara, Hiroyuki Sato, Zihao Zhu, Bo Zheng

In the film and gaming industries, achieving a realistic hair appearance typically involves the use of strands originating from the scalp. However, reconstructing these strands from observed surface images of hair presents significant challenges. The difficulty in acquiring Ground Truth (GT) data has led state-of-the-art learning-based methods to rely on pre-training with manually prepared synthetic CG data. This process is not only labor-intensive and costly but also introduces complications due to the domain gap when compared to real-world data. In this study, we propose an optimization-based approach that eliminates the need for pre-training. Our method represents hair strands as line segments growing from the scalp and optimizes them using a novel differentiable rendering algorithm. To robustly optimize a substantial number of slender explicit geometries, we introduce 3D orientation estimation utilizing global optimization, strand initialization based on Laplace's equation, and reparameterization that leverages geometric connectivity and spatial proximity. Unlike existing optimization-based methods, our method is capable of reconstructing internal hair flow in an absolute direction. Our method exhibits robust and accurate inverse rendering, surpassing the quality of existing methods and significantly improving processing speed.

comment: CVPR 2024

☆ Robust Containment Queries over Collections of Rational Parametric Curves via Generalized Winding Numbers

Jacob Spainhour, David Gunderman, Kenneth Weiss

Point containment queries for regions bound by watertight geometric surfaces, i.e. closed and without self-intersections, can be evaluated straightforwardly with a number of well-studied algorithms. However, when such assumptions on domain geometry are not met, these methods are theoretically unfounded at best and practically unusable at worst. More robust classification schemes utilize generalized winding numbers, a mathematical construction that is indifferent to imperfections in the often human-defined geometric model. We extend this methodology to more general curved shapes, defining a robust containment query for regions whose boundary elements are defined by a collection of rational parametric curves. In doing so, we devise an algorithm that is stable and accurate at arbitrary points in space, circumventing the typical difficulties for queries that are arbitrarily close or coincident with the model. This is done by reducing the generalized winding number problem to an integer winding number problem, which is solved by approximating each curve with a polyline that provably has the same winding number at the point of interest. We demonstrate the improvements in computational complexity granted by this method over conventional techniques, as well as the robustness induced by its application

comment: 14 pages, 18 figures

♻ ☆ Lodge: A Coarse to Fine Diffusion Network for Long Dance Generation Guided by the Characteristic Dance Primitives CVPR2024

Ronghui Li, YuXiang Zhang, Yachao Zhang, Hongwen Zhang, Jie Guo, Yan Zhang, Yebin Liu, Xiu Li

We propose Lodge, a network capable of generating extremely long dance sequences conditioned on given music. We design Lodge as a two-stage coarse to fine diffusion architecture, and propose the characteristic dance primitives that possess significant expressiveness as intermediate representations between two diffusion models. The first stage is global diffusion, which focuses on comprehending the coarse-level music-dance correlation and production characteristic dance primitives. In contrast, the second-stage is the local diffusion, which parallelly generates detailed motion sequences under the guidance of the dance primitives and choreographic rules. In addition, we propose a Foot Refine Block to optimize the contact between the feet and the ground, enhancing the physical realism of the motion. Our approach can parallelly generate dance sequences of extremely long length, striking a balance between global choreographic patterns and local motion quality and expressiveness. Extensive experiments validate the efficacy of our method.

comment: Accepted by CVPR2024, Project page: https://li-ronghui.github.io/lodge

♻ ☆ TRIPS: Trilinear Point Splatting for Real-Time Radiance Field Rendering

Linus Franke, Darius Rückert, Laura Fink, Marc Stamminger

Point-based radiance field rendering has demonstrated impressive results for novel view synthesis, offering a compelling blend of rendering quality and computational efficiency. However, also latest approaches in this domain are not without their shortcomings. 3D Gaussian Splatting [Kerbl and Kopanas et al. 2023] struggles when tasked with rendering highly detailed scenes, due to blurring and cloudy artifacts. On the other hand, ADOP [R\"uckert et al. 2022] can accommodate crisper images, but the neural reconstruction network decreases performance, it grapples with temporal instability and it is unable to effectively address large gaps in the point cloud. In this paper, we present TRIPS (Trilinear Point Splatting), an approach that combines ideas from both Gaussian Splatting and ADOP. The fundamental concept behind our novel technique involves rasterizing points into a screen-space image pyramid, with the selection of the pyramid layer determined by the projected point size. This approach allows rendering arbitrarily large points using a single trilinear write. A lightweight neural network is then used to reconstruct a hole-free image including detail beyond splat resolution. Importantly, our render pipeline is entirely differentiable, allowing for automatic optimization of both point sizes and positions. Our evaluation demonstrate that TRIPS surpasses existing state-of-the-art methods in terms of rendering quality while maintaining a real-time frame rate of 60 frames per second on readily available hardware. This performance extends to challenging scenarios, such as scenes featuring intricate geometry, expansive landscapes, and auto-exposed footage. The project page is located at: https://lfranke.github.io/trips/

♻ ☆ NeuS-PIR: Learning Relightable Neural Surface using Pre-Integrated Rendering

Shi Mao, Chenming Wu, Zhelun Shen, Yifan Wang, Dayan Wu, Liangjun Zhang

This paper presents a method, namely NeuS-PIR, for recovering relightable neural surfaces using pre-integrated rendering from multi-view images or video. Unlike methods based on NeRF and discrete meshes, our method utilizes implicit neural surface representation to reconstruct high-quality geometry, which facilitates the factorization of the radiance field into two components: a spatially varying material field and an all-frequency lighting representation. This factorization, jointly optimized using an adapted differentiable pre-integrated rendering framework with material encoding regularization, in turn addresses the ambiguity of geometry reconstruction and leads to better disentanglement and refinement of each scene property. Additionally, we introduced a method to distil indirect illumination fields from the learned representations, further recovering the complex illumination effect like inter-reflection. Consequently, our method enables advanced applications such as relighting, which can be seamlessly integrated with modern graphics engines. Qualitative and quantitative experiments have shown that NeuS-PIR outperforms existing methods across various tasks on both synthetic and real datasets. Source code is available at https://github.com/Sheldonmao/NeuSPIR

♻ ☆ Diffusion Models Generate Images Like Painters: an Analytical Theory of Outline First, Details Later NeurIPS23

Binxu Wang, John J. Vastola

How do diffusion generative models convert pure noise into meaningful images? In a variety of pretrained diffusion models (including conditional latent space models like Stable Diffusion), we observe that the reverse diffusion process that underlies image generation has the following properties: (i) individual trajectories tend to be low-dimensional and resemble 2D `rotations'; (ii) high-variance scene features like layout tend to emerge earlier, while low-variance details tend to emerge later; and (iii) early perturbations tend to have a greater impact on image content than later perturbations. To understand these phenomena, we derive and study a closed-form solution to the probability flow ODE for a Gaussian distribution, which shows that the reverse diffusion state rotates towards a gradually-specified target on the image manifold. It also shows that generation involves first committing to an outline, and then to finer and finer details. We find that this solution accurately describes the initial phase of image generation for pretrained models, and can in principle be used to make image generation more efficient by skipping reverse diffusion steps. Finally, we use our solution to characterize the image manifold in Stable Diffusion. Our viewpoint reveals an unexpected similarity between generation by GANs and diffusion and provides a conceptual link between diffusion and image retrieval.

comment: 44 pages, 28 figures. A briefer version was presented at NeurIPS23 Workshop on Diffusion Models [arXiv:2311.10892]

♻ ☆ Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models ICCV 2023

Songwei Ge, Seungjun Nah, Guilin Liu, Tyler Poon, Andrew Tao, Bryan Catanzaro, David Jacobs, Jia-Bin Huang, Ming-Yu Liu, Yogesh Balaji

Despite tremendous progress in generating high-quality images using diffusion models, synthesizing a sequence of animated frames that are both photorealistic and temporally coherent is still in its infancy. While off-the-shelf billion-scale datasets for image generation are available, collecting similar video data of the same scale is still challenging. Also, training a video diffusion model is computationally much more expensive than its image counterpart. In this work, we explore finetuning a pretrained image diffusion model with video data as a practical solution for the video synthesis task. We find that naively extending the image noise prior to video noise prior in video diffusion leads to sub-optimal performance. Our carefully designed video noise prior leads to substantially better performance. Extensive experimental validation shows that our model, Preserve Your Own Correlation (PYoCo), attains SOTA zero-shot text-to-video results on the UCF-101 and MSR-VTT benchmarks. It also achieves SOTA video generation quality on the small-scale UCF-101 benchmark with a $10\times$ smaller model using significantly less computation than the prior art.

comment: ICCV 2023. Project webpage: https://research.nvidia.com/labs/dir/pyoco

☆ A Robotic Skill Learning System Built Upon Diffusion Policies and Foundation Models

Nils Ingelhag, Jesper Munkeby, Jonne van Haastregt, Anastasia Varava, Michael C. Welle, Danica Kragic

In this paper, we build upon two major recent developments in the field, Diffusion Policies for visuomotor manipulation and large pre-trained multimodal foundational models to obtain a robotic skill learning system. The system can obtain new skills via the behavioral cloning approach of visuomotor diffusion policies given teleoperated demonstrations. Foundational models are being used to perform skill selection given the user's prompt in natural language. Before executing a skill the foundational model performs a precondition check given an observation of the workspace. We compare the performance of different foundational models to this end as well as give a detailed experimental evaluation of the skills taught by the user in simulation and the real world. Finally, we showcase the combined system on a challenging food serving scenario in the real world. Videos of all experimental executions, as well as the process of teaching new skills in simulation and the real world, are available on the project's website.

comment: https://roboskillframework.github.io

☆ BatDeck: Advancing Nano-drone Navigation with Low-power Ultrasound-based Obstacle Avoidance

Hanna Müller, Victor Kartsch, Michele Magno, Luca Benini

Nano-drones, distinguished by their agility, minimal weight, and cost-effectiveness, are particularly well-suited for exploration in confined, cluttered and narrow spaces. Recognizing transparent, highly reflective or absorbing materials, such as glass and metallic surfaces is challenging, as classical sensors, such as cameras or laser rangers, often do not detect them. Inspired by bats, which can fly at high speeds in complete darkness with the help of ultrasound, this paper introduces \textit{BatDeck}, a pioneering sensor-deck employing a lightweight and low-power ultrasonic sensor for nano-drone autonomous navigation. This paper first provides insights about sensor characteristics, highlighting the influence of motor noise on the ultrasound readings, then it introduces the results of extensive experimental tests for obstacle avoidance (OA) in a diverse environment. Results show that \textit{BatDeck} allows exploration for a flight time of 8 minutes while covering 136m on average before crash in a challenging environment with transparent and reflective obstacles, proving the effectiveness of ultrasonic sensors for OA on nano-drones.

comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

☆ Synapse: Learning Preferential Concepts from Visual Demonstrations

Sadanand Modak, Noah Patton, Isil Dillig, Joydeep Biswas

This paper addresses the problem of preference learning, which aims to learn user-specific preferences (e.g., "good parking spot", "convenient drop-off location") from visual input. Despite its similarity to learning factual concepts (e.g., "red cube"), preference learning is a fundamentally harder problem due to its subjective nature and the paucity of person-specific training data. We address this problem using a new framework called Synapse, which is a neuro-symbolic approach designed to efficiently learn preferential concepts from limited demonstrations. Synapse represents preferences as neuro-symbolic programs in a domain-specific language (DSL) that operates over images, and leverages a novel combination of visual parsing, large language models, and program synthesis to learn programs representing individual preferences. We evaluate Synapse through extensive experimentation including a user case study focusing on mobility-related concepts in mobile robotics and autonomous driving. Our evaluation demonstrates that Synapse significantly outperforms existing baselines as well as its own ablations. The code and other details can be found on the project website https://amrl.cs.utexas.edu/synapse .

comment: 23 pages, 7 figures; Preprint

☆ Domain Adaptive Detection of MAVs: A Benchmark and Noise Suppression Network

Yin Zhang, Jinhong Deng, Peidong Liu, Wen Li, Shiyu Zhao

Visual detection of Micro Air Vehicles (MAVs) has attracted increasing attention in recent years due to its important application in various tasks. The existing methods for MAV detection assume that the training set and testing set have the same distribution. As a result, when deployed in new domains, the detectors would have a significant performance degradation due to domain discrepancy. In this paper, we study the problem of cross-domain MAV detection. The contributions of this paper are threefold. 1) We propose a Multi-MAV-Multi-Domain (M3D) dataset consisting of both simulation and realistic images. Compared to other existing datasets, the proposed one is more comprehensive in the sense that it covers rich scenes, diverse MAV types, and various viewing angles. A new benchmark for cross-domain MAV detection is proposed based on the proposed dataset. 2) We propose a Noise Suppression Network (NSN) based on the framework of pseudo-labeling and a large-to-small training procedure. To reduce the challenging pseudo-label noises, two novel modules are designed in this network. The first is a prior-based curriculum learning module for allocating adaptive thresholds for pseudo labels with different difficulties. The second is a masked copy-paste augmentation module for pasting truly-labeled MAVs on unlabeled target images and thus decreasing pseudo-label noises. 3) Extensive experimental results verify the superior performance of the proposed method compared to the state-of-the-art ones. In particular, it achieves mAP of 46.9%(+5.8%), 50.5%(+3.7%), and 61.5%(+11.3%) on the tasks of simulation-to-real adaptation, cross-scene adaptation, and cross-camera adaptation, respectively.

comment: 17 pages, 11 figures. Accepted by IEEE Transactions on Automation Science and Engineering

☆ Skill Q-Network: Learning Adaptive Skill Ensemble for Mapless Navigation in Unknown Environments

Hyunki Seong, David Hyunchul Shim

This paper focuses on the acquisition of mapless navigation skills within unknown environments. We introduce the Skill Q-Network (SQN), a novel reinforcement learning method featuring an adaptive skill ensemble mechanism. Unlike existing methods, our model concurrently learns a high-level skill decision process alongside multiple low-level navigation skills, all without the need for prior knowledge. Leveraging a tailored reward function for mapless navigation, the SQN is capable of learning adaptive maneuvers that incorporate both exploration and goal-directed skills, enabling effective navigation in new environments. Our experiments demonstrate that our SQN can effectively navigate complex environments, exhibiting a 40% higher performance compared to baseline models. Without explicit guidance, SQN discovers how to combine low-level skill policies, showcasing both goal-directed navigations to reach destinations and exploration maneuvers to escape from local minimum regions in challenging scenarios. Remarkably, our adaptive skill ensemble method enables zero-shot transfer to out-of-distribution domains, characterized by unseen observations from non-convex obstacles or uneven, subterranean-like environments.

comment: 8 pages, 8 figures

☆ Trajectory Planning of Robotic Manipulator in Dynamic Environment Exploiting DRL

Osama Ahmad, Zawar Hussain, Hammad Naeem

This study is about the implementation of a reinforcement learning algorithm in the trajectory planning of manipulators. We have a 7-DOF robotic arm to pick and place the randomly placed block at a random target point in an unknown environment. The obstacle is randomly moving which creates a hurdle in picking the object. The objective of the robot is to avoid the obstacle and pick the block with constraints to a fixed timestamp. In this literature, we have applied a deep deterministic policy gradient (DDPG) algorithm and compared the model's efficiency with dense and sparse rewards.

comment: Accepted in ICIESTR-2024

☆ Bridging the Sim-to-Real Gap with Bayesian Inference

Jonas Rothfuss, Bhavya Sukhija, Lenart Treven, Florian Dörfler, Stelian Coros, Andreas Krause

We present SIM-FSVGD for learning robot dynamics from data. As opposed to traditional methods, SIM-FSVGD leverages low-fidelity physical priors, e.g., in the form of simulators, to regularize the training of neural network models. While learning accurate dynamics already in the low data regime, SIM-FSVGD scales and excels also when more data is available. We empirically show that learning with implicit physical priors results in accurate mean model estimation as well as precise uncertainty quantification. We demonstrate the effectiveness of SIM-FSVGD in bridging the sim-to-real gap on a high-performance RC racecar system. Using model-based RL, we demonstrate a highly dynamic parking maneuver with drifting, using less than half the data compared to the state of the art.

☆ Symbolic and User-friendly Geometric Algebra Routines (SUGAR) for Computations in Matlab

Manel Velasco, Isiah Zaplana, Arnau Dória-Cerezo, Pau Martí

Geometric algebra (GA) is a mathematical tool for geometric computing, providing a framework that allows a unified and compact approach to geometric relations which in other mathematical systems are typically described using different more complicated elements. This fact has led to an increasing adoption of GA in applied mathematics and engineering problems. However, the scarcity of symbolic implementations of GA and its inherent complexity, requiring a specific mathematical background, make it challenging and less intuitive for engineers to work with. This prevents wider adoption among more applied professionals. To address this challenge, this paper introduces SUGAR (Symbolic and User-friendly Geometric Algebra Routines), an open-source toolbox designed for Matlab and licensed under the MIT License. SUGAR facilitates the translation of GA concepts into Matlab and provides a collection of user-friendly functions tailored for GA computations, including support for symbolic operations. It supports both numeric and symbolic computations in high-dimensional GAs. Specifically tailored for applied mathematics and engineering applications, SUGAR has been meticulously engineered to represent geometric elements and transformations within two and three-dimensional projective and conformal geometric algebras, aligning with established computational methodologies in the literature. Furthermore, SUGAR efficiently handles functions of multivectors, such as exponential, logarithmic, sinusoidal, and cosine functions, enhancing its applicability across various engineering domains, including robotics, control systems, and power electronics. Finally, this work includes four distinct validation examples, demonstrating SUGAR's capabilities across the above-mentioned fields and its practical utility in addressing real-world applied mathematics and engineering problems.

comment: 33 pages, 6 figures, journal paper submitted to ACM TOMS

☆ Technical Development of a Semi-Autonomous Robotic Partition

Binh Vinh Duc Nguyen, Andrew Vande Moere

This technical description details the design and engineering process of a semi-autonomous robotic partition. This robotic partition prototype was subsequently employed in a longer-term evaluation in-the-wild study conducted by the authors in a real-world office setting.

☆ ROXIE: Defining a Robotic eXplanation and Interpretability Engine IROS 2024

Francisco J. Rodríguez-Lera, Miguel A. González-Santamarta, Alejandro González-Cantón, Laura Fernández-Becerra, David Sobrín-Hidalgo, Angel Manuel Guerrero-Higueras

In an era where autonomous robots increasingly inhabit public spaces, the imperative for transparency and interpretability in their decision-making processes becomes paramount. This paper presents the overview of a Robotic eXplanation and Interpretability Engine (ROXIE), which addresses this critical need, aiming to demystify the opaque nature of complex robotic behaviors. This paper elucidates the key features and requirements needed for providing information and explanations about robot decision-making processes. It also overviews the suite of software components and libraries available for deployment with ROS 2, empowering users to provide comprehensive explanations and interpretations of robot processes and behaviors, thereby fostering trust and collaboration in human-robot interactions.

comment: 7 pages, 3 figures, 1 tables, Submitted to IROS 2024

☆ Research Challenges for Adaptive Architecture: Empowering Occupants of Multi-Occupancy Buildings

Binh Vinh Duc Nguyen, Andrew Vande Moere

This positional paper outlines our vision of 'adaptive architecture', which involves the integration of robotic technology to physically change an architectural space in supporting the changing needs of its occupants, in response to the CHI'24 workshop "HabiTech - Inhabiting Buildings, Data & Technology" call on "How do new technologies enable and empower the inhabitants of multi-occupancy buildings?". Specifically, while adaptive architecture holds promise for enhancing occupant satisfaction, comfort, and overall health and well-being, there remains a range of research challenges of (1) how it can effectively support individual occupants, while (2) mediating the conflicting needs of collocated others, and (3) integrating meaningfully into the sociocultural characteristics of their building community.

☆ The Adaptive Workplace: Orchestrating Architectural Services around the Wellbeing of Individual Occupants

Andrew Vande Moere, Sara Arko, Alena Safrova Drasilova, Tomáš Ondráček, Ilaria Pigliautile, Benedetta Pioppi, Anna Laura Pisello, Jakub Prochazka, Paula Acuna Roncancio, Davide Schaumann, Marcel Schweiker, Binh Vinh Duc Nguyen

As the academic consortia members of the EU Horizon project SONATA ("Situation-aware OrchestratioN of AdapTive Architecture"), we respond to the workshop call for "Office Wellbeing by Design: Don't Stand for Anything Less" by proposing the "Adaptive Workplace" concept. In essence, our vision aims to adapt a workplace to the ever-changing needs of individual occupants, instead of that occupants are expected to adapt to their workplace.

☆ Counter-example guided Imitation Learning of Feedback Controllers from Temporal Logic Specifications

Thao Dang, Alexandre Donzé, Inzemamul Haque, Nikolaos Kekatos, Indranil Saha

We present a novel method for imitation learning for control requirements expressed using Signal Temporal Logic (STL). More concretely we focus on the problem of training a neural network to imitate a complex controller. The learning process is guided by efficient data aggregation based on counter-examples and a coverage measure. Moreover, we introduce a method to evaluate the performance of the learned controller via parameterization and parameter estimation of the STL requirements. We demonstrate our approach with a flying robot case study.

☆ Active Admittance Control with Iterative Learning for General-Purpose Contact-Rich Manipulation

Bo Zhou, Yuyao Sun, Wenbo Liu, Ruixuan Jiao, Fang Fang, Shihua Li

Force interaction is inevitable when robots face multiple operation scenarios. How to make the robot competent in force control for generalized operations such as multi-tasks still remains a challenging problem. Aiming at the reproducibility of interaction tasks and the lack of a generalized force control framework for multi-task scenarios, this paper proposes a novel hybrid control framework based on active admittance control with iterative learning parameters-tunning mechanism. The method adopts admittance control as the underlying algorithm to ensure flexibility, and iterative learning as the high-level algorithm to regulate the parameters of the admittance model. The whole algorithm has flexibility and learning ability, which is capable of achieving the goal of excellent versatility. Four representative interactive robot manipulation tasks are chosen to investigate the consistency and generalisability of the proposed method. Experiments are designed to verify the effectiveness of the whole framework, and an average of 98.21% and 91.52% improvement of RMSE is obtained relative to the traditional admittance control as well as the model-free adaptive control, respectively.

☆ Arm-Constrained Curriculum Learning for Loco-Manipulation of the Wheel-Legged Robot

Zifan Wang, Yufei Jia, Lu Shi, Haoyu Wang, Haizhou Zhao, Xueyang Li, Jinni Zhou, Jun Ma, Guyue Zhou

Incorporating a robotic manipulator into a wheel-legged robot enhances its agility and expands its potential for practical applications. However, the presence of potential instability and uncertainties presents additional challenges for control objectives. In this paper, we introduce an arm-constrained curriculum learning architecture to tackle the issues introduced by adding the manipulator. Firstly, we develop an arm-constrained reinforcement learning algorithm to ensure safety and stability in control performance. Additionally, to address discrepancies in reward settings between the arm and the base, we propose a reward-aware curriculum learning method. The policy is first trained in Isaac gym and transferred to the physical robot to do dynamic grasping tasks, including the door-opening task, fan-twitching task and the relay-baton-picking and following task. The results demonstrate that our proposed approach effectively controls the arm-equipped wheel-legged robot to master dynamic grasping skills, allowing it to chase and catch a moving object while in motion. The code can be found at https://github.com/aCodeDog/legged-robots-manipulation. To view the supplemental video, please visit https://youtu.be/sNXT-rwPNMM.

☆ Hallucination Detection in Foundation Models for Decision-Making: A Flexible Definition and Review of the State of the Art

Neeloy Chakraborty, Melkior Ornik, Katherine Driggs-Campbell

Autonomous systems are soon to be ubiquitous, from manufacturing autonomy to agricultural field robots, and from health care assistants to the entertainment industry. The majority of these systems are developed with modular sub-components for decision-making, planning, and control that may be hand-engineered or learning-based. While these existing approaches have been shown to perform well under the situations they were specifically designed for, they can perform especially poorly in rare, out-of-distribution scenarios that will undoubtedly arise at test-time. The rise of foundation models trained on multiple tasks with impressively large datasets from a variety of fields has led researchers to believe that these models may provide common sense reasoning that existing planners are missing. Researchers posit that this common sense reasoning will bridge the gap between algorithm development and deployment to out-of-distribution tasks, like how humans adapt to unexpected scenarios. Large language models have already penetrated the robotics and autonomous systems domains as researchers are scrambling to showcase their potential use cases in deployment. While this application direction is very promising empirically, foundation models are known to hallucinate and generate decisions that may sound reasonable, but are in fact poor. We argue there is a need to step back and simultaneously design systems that can quantify the certainty of a model's decision, and detect when it may be hallucinating. In this work, we discuss the current use cases of foundation models for decision-making tasks, provide a general definition for hallucinations with examples, discuss existing approaches to hallucination detection and mitigation with a focus on decision problems, and explore areas for further research in this exciting field.

comment: 31 pages, 2 tables

☆ Spatially temporally distributed informative path planning for multi-robot systems

Binh Nguyen, Linh Nguyen, Truong X. Nghiem, Hung La, Jose Baca, Pablo Rangel, Miguel Cid Montoya, Thang Nguyen

This paper investigates the problem of informative path planning for a mobile robotic sensor network in spatially temporally distributed mapping. The robots are able to gather noisy measurements from an area of interest during their movements to build a Gaussian Process (GP) model of a spatio-temporal field. The model is then utilized to predict the spatio-temporal phenomenon at different points of interest. To spatially and temporally navigate the group of robots so that they can optimally acquire maximal information gains while their connectivity is preserved, we propose a novel multistep prediction informative path planning optimization strategy employing our newly defined local cost functions. By using the dual decomposition method, it is feasible and practical to effectively solve the optimization problem in a distributed manner. The proposed method was validated through synthetic experiments utilizing real-world data sets.

☆ Real-time Model Predictive Control with Zonotope-Based Neural Networks for Bipedal Social Navigation

Abdulaziz Shamsah, Krishanu Agarwal, Shreyas Kousik, Ye Zhao

This study addresses the challenge of bipedal navigation in a dynamic human-crowded environment, a research area that remains largely underexplored in the field of legged navigation. We propose two cascaded zonotope-based neural networks: a Pedestrian Prediction Network (PPN) for pedestrians' future trajectory prediction and an Ego-agent Social Network (ESN) for ego-agent social path planning. Representing future paths as zonotopes allows for efficient reachability-based planning and collision checking. The ESN is then integrated with a Model Predictive Controller (ESN-MPC) for footstep planning for our bipedal robot Digit designed by Agility Robotics. ESN-MPC solves for a collision-free optimal trajectory by optimizing through the gradients of ESN. ESN-MPC optimal trajectory is sent to the low-level controller for full-order simulation of Digit. The overall proposed framework is validated with extensive simulations on randomly generated initial settings with varying human crowd densities.

comment: 8 pages, 9 figures

☆ Towards Cooperative Maneuver Planning in Mixed Traffic at Urban Intersections

Marvin Klimke, Max Bastian Mertens, Benjamin Völz, Michael Buchholz

Connected automated driving promises a significant improvement of traffic efficiency and safety on highways and in urban areas. Apart from sharing of awareness and perception information over wireless communication links, cooperative maneuver planning may facilitate active guidance of connected automated vehicles at urban intersections. Research in automatic intersection management put forth a large body of works that mostly employ rule-based or optimization-based approaches primarily in fully automated simulated environments. In this work, we present two cooperative planning approaches that are capable of handling mixed traffic, i.e., the road being shared by automated vehicles and regular vehicles driven by humans. Firstly, we propose an optimization-based planner trained on real driving data that cyclically selects the most efficient out of multiple predicted coordinated maneuvers. Additionally, we present a cooperative planning approach based on graph-based reinforcement learning, which conquers the lack of ground truth data for cooperative maneuvers. We present evaluation results of both cooperative planners in high-fidelity simulation and real-world traffic. Simulative experiments in fully automated traffic and mixed traffic show that cooperative maneuver planning leads to less delay due to interaction and a reduced number of stops. In real-world experiments with three prototype connected automated vehicles in public traffic, both planners demonstrate their ability to perform efficient cooperative maneuvers.

comment: M. Klimke and M. Mertens are both first authors with equal contribution. 11 pages, 10 figures, 2 tables, submitted to IEEE Transactions on Intelligent Vehicles

☆ Producing and Leveraging Online Map Uncertainty in Trajectory Prediction CVPR 2024

Xunjiang Gu, Guanyu Song, Igor Gilitschenski, Marco Pavone, Boris Ivanovic

High-definition (HD) maps have played an integral role in the development of modern autonomous vehicle (AV) stacks, albeit with high associated labeling and maintenance costs. As a result, many recent works have proposed methods for estimating HD maps online from sensor data, enabling AVs to operate outside of previously-mapped regions. However, current online map estimation approaches are developed in isolation of their downstream tasks, complicating their integration in AV stacks. In particular, they do not produce uncertainty or confidence estimates. In this work, we extend multiple state-of-the-art online map estimation methods to additionally estimate uncertainty and show how this enables more tightly integrating online mapping with trajectory forecasting. In doing so, we find that incorporating uncertainty yields up to 50% faster training convergence and up to 15% better prediction performance on the real-world nuScenes driving dataset.

comment: 14 pages, 14 figures, 6 tables. CVPR 2024

☆ AeroBridge: Autonomous Drone Handoff System for Emergency Battery Service

Avishkar Seth, Alice James, Endrowednes Kuantama, Richard Han, Subhas Mukhopadhyay

This paper proposes an Emergency Battery Service (EBS) for drones in which an EBS drone flies to a drone in the field with a depleted battery and transfers a fresh battery to the exhausted drone. The authors present a unique battery transfer mechanism and drone localization that uses the Cross Marker Position (CMP) method. The main challenges include a stable and balanced transfer that precisely localizes the receiver drone. The proposed EBS drone mitigates the effects of downwash due to the vertical proximity between the drones by implementing diagonal alignment with the receiver, reducing the distance to 0.5 m between the two drones. CFD analysis shows that diagonal instead of perpendicular alignment minimizes turbulence, and the authors verify the actual system for change in output airflow and thrust measurements. The CMP marker-based localization method enables position lock for the EBS drone with up to 0.9 cm accuracy. The performance of the transfer mechanism is validated experimentally by successful mid-air transfer in 5 seconds, where the EBS drone is within 0.5 m vertical distance from the receiver drone, wherein 4m/s turbulence does not affect the transfer process.

☆ Enhancing Visual Place Recognition via Fast and Slow Adaptive Biasing in Event Cameras

Gokul B. Nair, Michael Milford, Tobias Fischer

Event cameras are increasingly popular in robotics due to their beneficial features, such as low latency, energy efficiency, and high dynamic range. Nevertheless, their downstream task performance is greatly influenced by the optimization of bias parameters. These parameters, for instance, regulate the necessary change in light intensity to trigger an event, which in turn depends on factors such as the environment lighting and camera motion. This paper introduces feedback control algorithms that automatically tune the bias parameters through two interacting methods: 1) An immediate, on-the-fly fast adaptation of the refractory period, which sets the minimum interval between consecutive events, and 2) if the event rate exceeds the specified bounds even after changing the refractory period repeatedly, the controller adapts the pixel bandwidth and event thresholds, which stabilizes after a short period of noise events across all pixels (slow adaptation). Our evaluation focuses on the visual place recognition task, where incoming query images are compared to a given reference database. We conducted comprehensive evaluations of our algorithms' adaptive feedback control in real-time. To do so, we collected the QCR-Fast-and-Slow dataset that contains DAVIS346 event camera streams from 366 repeated traversals of a Scout Mini robot navigating through a 100 meter long indoor lab setting (totaling over 35km distance traveled) in varying brightness conditions with ground truth location information. Our proposed feedback controllers result in superior performance when compared to the standard bias settings and prior feedback control methods. Our findings also detail the impact of bias adjustments on task performance and feature ablation studies on the fast and slow adaptation mechanisms.

comment: 8 pages, 9 figures, paper under review

☆ Terrain-Attentive Learning for Efficient 6-DoF Kinodynamic Modeling on Vertically Challenging Terrain

Aniket Datar, Chenhui Pan, Mohammad Nazeri, Anuj Pokhrel, Xuesu Xiao

Wheeled robots have recently demonstrated superior mechanical capability to traverse vertically challenging terrain (e.g., extremely rugged boulders comparable in size to the vehicles themselves). Negotiating such terrain introduces significant variations of vehicle pose in all six Degrees-of-Freedom (DoFs), leading to imbalanced contact forces, varying momentum, and chassis deformation due to non-rigid tires and suspensions. To autonomously navigate on vertically challenging terrain, all these factors need to be efficiently reasoned within limited onboard computation and strict real-time constraints. In this paper, we propose a 6-DoF kinodynamics learning approach that is attentive only to the specific underlying terrain critical to the current vehicle-terrain interaction, so that it can be efficiently queried in real-time motion planners onboard small robots. Physical experiment results show our Terrain-Attentive Learning demonstrates on average 51.1% reduction in model prediction error among all 6 DoFs compared to a state-of-the-art model for vertically challenging terrain.

☆ ASDF: Assembly State Detection Utilizing Late Fusion by Integrating 6D Pose Estimation

Hannah Schieber, Shiyu Li, Niklas Corell, Philipp Beckerle, Julian Kreimeier, Daniel Roth

In medical and industrial domains, providing guidance for assembly processes is critical to ensure efficiency and safety. Errors in assembly can lead to significant consequences such as extended surgery times, and prolonged manufacturing or maintenance times in industry. Assembly scenarios can benefit from in-situ AR visualization to provide guidance, reduce assembly times and minimize errors. To enable in-situ visualization 6D pose estimation can be leveraged. Existing 6D pose estimation techniques primarily focus on individual objects and static captures. However, assembly scenarios have various dynamics including occlusion during assembly and dynamics in the assembly objects appearance. Existing work, combining object detection/6D pose estimation and assembly state detection focuses either on pure deep learning-based approaches, or limit the assembly state detection to building blocks. To address the challenges of 6D pose estimation in combination with assembly state detection, our approach ASDF builds upon the strengths of YOLOv8, a real-time capable object detection framework. We extend this framework, refine the object pose and fuse pose knowledge with network-detected pose information. Utilizing our late fusion in our Pose2State module results in refined 6D pose estimation and assembly state detection. By combining both pose and state information, our Pose2State module predicts the final assembly state with precision. Our evaluation on our ASDF dataset shows that our Pose2State module leads to an improved assembly state detection and that the improvement of the assembly state further leads to a more robust 6D pose estimation. Moreover, on the GBOT dataset, we outperform the pure deep learning-based network, and even outperform the hybrid and pure tracking-based approaches.

☆ SE(3) Linear Parameter Varying Dynamical Systems for Globally Asymptotically Stable End-Effector Control

Sunan Sun, Nadia Figueroa

Linear Parameter Varying Dynamical Systems (LPV-DS) encode trajectories into an autonomous first-order DS that enables reactive responses to perturbations, while ensuring globally asymptotic stability at the target. However, the current LPV-DS framework is established on Euclidean data only and has not been applicable to broader robotic applications requiring pose control. In this paper we present an extension to the current LPV-DS framework, named Quaternion-DS, which efficiently learns a DS-based motion policy for orientation. Leveraging techniques from differential geometry and Riemannian statistics, our approach properly handles the non-Euclidean orientation data in quaternion space, enabling the integration with positional control, namely SE(3) LPV-DS, so that the synergistic behaviour within the full SE(3) pose is preserved. Through simulation and real robot experiments, we validate our method, demonstrating its ability to efficiently and accurately reproduce the original SE(3) trajectory while exhibiting strong robustness to perturbations in task space.

☆ Bipedal Safe Navigation over Uncertain Rough Terrain: Unifying Terrain Mapping and Locomotion Stability

Kasidit Muenprasitivej, Jesse Jiang, Abdulaziz Shamsah, Samuel Coogan, Ye Zhao

We study the problem of bipedal robot navigation in complex environments with uncertain and rough terrain. In particular, we consider a scenario in which the robot is expected to reach a desired goal location by traversing an environment with uncertain terrain elevation. Such terrain uncertainties induce not only untraversable regions but also robot motion perturbations. Thus, the problems of terrain mapping and locomotion stability are intertwined. We evaluate three different kernels for Gaussian process (GP) regression to learn the terrain elevation. We also learn the motion deviation resulting from both the terrain as well as the discrepancy between the reduced-order Prismatic Inverted Pendulum Model used for planning and the full-order locomotion dynamics. We propose a hierarchical locomotion-dynamics-aware sampling-based navigation planner. The global navigation planner plans a series of local waypoints to reach the desired goal locations while respecting locomotion stability constraints. Then, a local navigation planner is used to generate a sequence of dynamically feasible footsteps to reach local waypoints. We develop a novel trajectory evaluation metric to minimize motion deviation and maximize information gain of the terrain elevation map. We evaluate the efficacy of our planning framework on Digit bipedal robot simulation in MuJoCo.

comment: 9 pages, 7 figures

☆ Human Stress Response and Perceived Safety during Encounters with Quadruped Robots

Ryan Gupta, Hyonyoung Shin, Emily Norman, Keri K. Stephens, Nanshu Lu, Luis Sentis

Despite the rise of mobile robot deployments in home and work settings, perceived safety of users and bystanders is understudied in the human-robot interaction (HRI) literature. To address this, we present a study designed to identify elements of a human-robot encounter that correlate with observed stress response. Stress is a key component of perceived safety and is strongly associated with human physiological response. In this study a Boston Dynamics Spot and a Unitree Go1 navigate autonomously through a shared environment occupied by human participants wearing multimodal physiological sensors to track their electrocardiography (ECG) and electrodermal activity (EDA). The encounters are varied through several trials and participants self-rate their stress levels after each encounter. The study resulted in a multidimensional dataset archiving various objective and subjective aspects of a human-robot encounter, containing insights for understanding perceived safety in such encounters. To this end, acute stress responses were decoded from the human participants' ECG and EDA and compared across different human-robot encounter conditions. Statistical analysis of data indicate that on average (1) participants feel more stress during encounters compared to baselines, (2) participants feel more stress encountering multiple robots compared to a single robot and (3) participants stress increases during navigation behavior compared with search behavior.

comment: 7 pages, 7 figs, 5 tables

☆ Exploring CausalWorld: Enhancing robotic manipulation via knowledge transfer and curriculum learning

Xinrui Wang, Yan Jin

This study explores a learning-based tri-finger robotic arm manipulating task, which requires complex movements and coordination among the fingers. By employing reinforcement learning, we train an agent to acquire the necessary skills for proficient manipulation. To enhance the efficiency and effectiveness of the learning process, two knowledge transfer strategies, fine-tuning and curriculum learning, were utilized within the soft actor-critic architecture. Fine-tuning allows the agent to leverage pre-trained knowledge and adapt it to new tasks. Several variations like model transfer, policy transfer, and across-task transfer were implemented and evaluated. To eliminate the need for pretraining, curriculum learning decomposes the advanced task into simpler, progressive stages, mirroring how humans learn. The number of learning stages, the context of the sub-tasks, and the transition timing were found to be the critical design parameters. The key factors of two learning strategies and corresponding effects were explored in context-aware and context-unaware scenarios, enabling us to identify the scenarios where the methods demonstrate optimal performance, derive conclusive insights, and contribute to a broader range of learning-based engineering applications.

☆ Impact-Aware Bimanual Catching of Large-Momentum Objects

Lei Yan, Theodoros Stouraitis, João Moura, Wenfu Xu, Michael Gienger, Sethu Vijayakumar

This paper investigates one of the most challenging tasks in dynamic manipulation -- catching large-momentum moving objects. Beyond the realm of quasi-static manipulation, dealing with highly dynamic objects can significantly improve the robot's capability of interacting with its surrounding environment. Yet, the inevitable motion mismatch between the fast moving object and the approaching robot will result in large impulsive forces, which lead to the unstable contacts and irreversible damage to both the object and the robot. To address the above problems, we propose an online optimization framework to: 1) estimate and predict the linear and angular motion of the object; 2) search and select the optimal contact locations across every surface of the object to mitigate impact through sequential quadratic programming (SQP); 3) simultaneously optimize the end-effector motion, stiffness, and contact force for both robots using multi-mode trajectory optimization (MMTO); and 4) realise the impact-aware catching motion on the compliant robotic system based on indirect force controller. We validate the impulse distribution, contact selection, and impact-aware MMTO algorithms in simulation and demonstrate the benefits of the proposed framework in real-world experiments including catching large-momentum moving objects with well-defined motion, constrained motion and free-flying motion.

☆ DASA: Delay-Adaptive Multi-Agent Stochastic Approximation

Nicolo Dal Fabbro, Arman Adibi, H. Vincent Poor, Sanjeev R. Kulkarni, Aritra Mitra, George J. Pappas

We consider a setting in which $N$ agents aim to speedup a common Stochastic Approximation (SA) problem by acting in parallel and communicating with a central server. We assume that the up-link transmissions to the server are subject to asynchronous and potentially unbounded time-varying delays. To mitigate the effect of delays and stragglers while reaping the benefits of distributed computation, we propose \texttt{DASA}, a Delay-Adaptive algorithm for multi-agent Stochastic Approximation. We provide a finite-time analysis of \texttt{DASA} assuming that the agents' stochastic observation processes are independent Markov chains. Significantly advancing existing results, \texttt{DASA} is the first algorithm whose convergence rate depends only on the mixing time $\tmix$ and on the average delay $\tau_{avg}$ while jointly achieving an $N$-fold convergence speedup under Markovian sampling. Our work is relevant for various SA applications, including multi-agent and distributed temporal difference (TD) learning, Q-learning and stochastic optimization with correlated data.

☆ TwoStep: Multi-agent Task Planning using Classical Planners and Large Language Models

Ishika Singh, David Traum, Jesse Thomason

Classical planning formulations like the Planning Domain Definition Language (PDDL) admit action sequences guaranteed to achieve a goal state given an initial state if any are possible. However, reasoning problems defined in PDDL do not capture temporal aspects of action taking, for example that two agents in the domain can execute an action simultaneously if postconditions of each do not interfere with preconditions of the other. A human expert can decompose a goal into largely independent constituent parts and assign each agent to one of these subgoals to take advantage of simultaneous actions for faster execution of plan steps, each using only single agent planning. By contrast, large language models (LLMs) used for directly inferring plan steps do not guarantee execution success, but do leverage commonsense reasoning to assemble action sequences. We combine the strengths of classical planning and LLMs by approximating human intuitions for two-agent planning goal decomposition. We demonstrate that LLM-based goal decomposition leads to faster planning times than solving multi-agent PDDL problems directly while simultaneously achieving fewer plan execution steps than a single agent plan alone and preserving execution success. Additionally, we find that LLM-based approximations of subgoals can achieve similar multi-agent execution steps than those specified by human experts. Website and resources at https://glamor-usc.github.io/twostep

comment: 12 pages

☆ Temporal and Semantic Evaluation Metrics for Foundation Models in Post-Hoc Analysis of Robotic Sub-tasks IROS 2024

Jonathan Salfity, Selma Wanna, Minkyu Choi, Mitch Pryor

Recent works in Task and Motion Planning (TAMP) show that training control policies on language-supervised robot trajectories with quality labeled data markedly improves agent task success rates. However, the scarcity of such data presents a significant hurdle to extending these methods to general use cases. To address this concern, we present an automated framework to decompose trajectory data into temporally bounded and natural language-based descriptive sub-tasks by leveraging recent prompting strategies for Foundation Models (FMs) including both Large Language Models (LLMs) and Vision Language Models (VLMs). Our framework provides both time-based and language-based descriptions for lower-level sub-tasks that comprise full trajectories. To rigorously evaluate the quality of our automatic labeling framework, we contribute an algorithm SIMILARITY to produce two novel metrics, temporal similarity and semantic similarity. The metrics measure the temporal alignment and semantic fidelity of language descriptions between two sub-task decompositions, namely an FM sub-task decomposition prediction and a ground-truth sub-task decomposition. We present scores for temporal similarity and semantic similarity above 90%, compared to 30% of a randomized baseline, for multiple robotic environments, demonstrating the effectiveness of our proposed framework. Our results enable building diverse, large-scale, language-supervised datasets for improved robotic TAMP.

comment: 8 pages, 3 figures. IROS 2024 Submission

☆ Speeding Up Path Planning via Reinforcement Learning in MCTS for Automated Parking

Xinlong Zheng, Xiaozhou Zhang, Donghao Xu

In this paper, we address a method that integrates reinforcement learning into the Monte Carlo tree search to boost online path planning under fully observable environments for automated parking tasks. Sampling-based planning methods under high-dimensional space can be computationally expensive and time-consuming. State evaluation methods are useful by leveraging the prior knowledge into the search steps, making the process faster in a real-time system. Given the fact that automated parking tasks are often executed under complex environments, a solid but lightweight heuristic guidance is challenging to compose in a traditional analytical way. To overcome this limitation, we propose a reinforcement learning pipeline with a Monte Carlo tree search under the path planning framework. By iteratively learning the value of a state and the best action among samples from its previous cycle's outcomes, we are able to model a value estimator and a policy generator for given states. By doing that, we build up a balancing mechanism between exploration and exploitation, speeding up the path planning process while maintaining its quality without using human expert driver data.

☆ PROSPECT: Precision Robot Spectroscopy Exploration and Characterization Tool

Nathaniel Hanson, Gary Lvov, Vedant Rautela, Samuel Hibbard, Ethan Holand, Charles DiMarzio, Taşkın Padır

Near Infrared (NIR) spectroscopy is widely used in industrial quality control and automation to test the purity and material quality of items. In this research, we propose a novel sensorized end effector and acquisition strategy to capture spectral signatures from objects and register them with a 3D point cloud. Our methodology first takes a 3D scan of an object generated by a time-of-flight depth camera and decomposes the object into a series of planned viewpoints covering the surface. We generate motion plans for a robot manipulator and end-effector to visit these viewpoints while maintaining a fixed distance and surface normal to ensure maximal spectral signal quality enabled by the spherical motion of the end-effector. By continuously acquiring surface reflectance values as the end-effector scans the target object, the autonomous system develops a four-dimensional model of the target object: position in an R^3 coordinate frame, and a wavelength vector denoting the associated spectral signature. We demonstrate this system in building spectral-spatial object profiles of increasingly complex geometries. As a point of comparison, we show our proposed system and spectral acquisition planning yields more consistent signal signals than naive point scanning strategies for capturing spectral information over complex surface geometries. Our work represents a significant step towards high-resolution spectral-spatial sensor fusion for automated quality assessment.

☆ Dyna-LfLH: Learning Agile Navigation in Dynamic Environments from Learned Hallucination IROS

Saad Abdul Ghani, Zizhao Wang, Peter Stone, Xuesu Xiao

This paper presents a self-supervised learning method to safely learn a motion planner for ground robots to navigate environments with dense and dynamic obstacles. When facing highly-cluttered, fast-moving, hard-to-predict obstacles, classical motion planners may not be able to keep up with limited onboard computation. For learning-based planners, high-quality demonstrations are difficult to acquire for imitation learning while reinforcement learning becomes inefficient due to the high probability of collision during exploration. To safely and efficiently provide training data, the Learning from Hallucination (LfH) approaches synthesize difficult navigation environments based on past successful navigation experiences in relatively easy or completely open ones, but unfortunately cannot address dynamic obstacles. In our new Dynamic Learning from Learned Hallucination (Dyna-LfLH), we design and learn a novel latent distribution and sample dynamic obstacles from it, so the generated training data can be used to learn a motion planner to navigate in dynamic environments. Dyna-LfLH is evaluated on a ground robot in both simulated and physical environments and achieves up to 25% better success rate compared to baselines.

comment: Submitted to International Conference on Intelligent Robots and Systems (IROS) 2024

☆ Multi-Contact Inertial Estimation and Localization in Legged Robots

Sergi Martinez, Robert Griffin, Carlos Mastalli

Optimal estimation is a promising tool for multi-contact inertial estimation and localization. To harness its advantages in robotics, it is crucial to solve these large and challenging optimization problems efficiently. To tackle this, we (i) develop a multiple-shooting solver that exploits both temporal and parametric structures through a parametrized Riccati recursion. Additionally, we (ii) propose an inertial local manifold that ensures its full physical consistency. It also enhances convergence compared to the singularity-free log-Cholesky approach. To handle its singularities, we (iii) introduce a nullspace approach in our optimal estimation solver. We (iv) finally develop the analytical derivatives of contact dynamics for both inertial parametrizations. Our framework can successfully solve estimation problems for complex maneuvers such as brachiation in humanoids. We demonstrate its numerical capabilities across various robotics tasks and its benefits in experimental trials with the Go1 robot.

☆ Hearing the shape of an arena with spectral swarm robotics

Leo Cazenille, Nicolas Lobato-Dauzier, Alessia Loi, Mika Ito, Olivier Marchal, Nathanael Aubert-Kato, Nicolas Bredeche, Anthony J. Genot

Swarm robotics promises adaptability to unknown situations and robustness against failures. However, it still struggles with global tasks that require understanding the broader context in which the robots operate, such as identifying the shape of the arena in which the robots are embedded. Biological swarms, such as shoals of fish, flocks of birds, and colonies of insects, routinely solve global geometrical problems through the diffusion of local cues. This paradigm can be explicitly described by mathematical models that could be directly computed and exploited by a robotic swarm. Diffusion over a domain is mathematically encapsulated by the Laplacian, a linear operator that measures the local curvature of a function. Crucially the geometry of a domain can generally be reconstructed from the eigenspectrum of its Laplacian. Here we introduce spectral swarm robotics where robots diffuse information to their neighbors to emulate the Laplacian operator - enabling them to "hear" the spectrum of their arena. We reveal a universal scaling that links the optimal number of robots (a global parameter) with their optimal radius of interaction (a local parameter). We validate experimentally spectral swarm robotics under challenging conditions with the one-shot classification of arena shapes using a sparse swarm of Kilobots. Spectral methods can assist with challenging tasks where robots need to build an emergent consensus on their environment, such as adaptation to unknown terrains, division of labor, or quorum sensing. Spectral methods may extend beyond robotics to analyze and coordinate swarms of agents of various natures, such as traffic or crowds, and to better understand the long-range dynamics of natural systems emerging from short-range interactions.

☆ Adaptive Step Duration for Precise Foot Placement: Achieving Robust Bipedal Locomotion on Terrains with Restricted Footholds

Zhaoyang Xiang, Victor Paredes, Ayonga Hereid

This paper introduces a novel multi-step preview foot placement planning algorithm designed to enhance the robustness of bipedal robotic walking across challenging terrains with restricted footholds. Traditional one-step preview planning struggles to maintain stability when stepping areas are severely limited, such as with random stepping stones. In this work, we developed a discrete-time Model Predictive Control (MPC) based on the step-to-step discrete evolution of the Divergent Component of Motion (DCM) of bipedal locomotion. This approach adaptively changes the step duration for optimal foot placement under constraints, thereby ensuring the robot's operational viability over multiple future steps and significantly improving its ability to navigate through environments with tight constraints on possible footholds. The effectiveness of this planning algorithm is demonstrated through simulations that include a variety of complex stepping-stone configurations and external perturbations. These tests underscore the algorithm's improved performance for navigating foothold-restricted environments, even with the presence of external disturbances.

comment: 8 pages, 8 figures, submitted to CDC 2024, for associated simulation video, see https://youtu.be/2jhikPlZmbE

☆ Grounding Language Plans in Demonstrations Through Counterfactual Perturbations

Yanwei Wang, Tsun-Hsuan Wang, Jiayuan Mao, Michael Hagenow, Julie Shah

Grounding the common-sense reasoning of Large Language Models in physical domains remains a pivotal yet unsolved problem for embodied AI. Whereas prior works have focused on leveraging LLMs directly for planning in symbolic spaces, this work uses LLMs to guide the search of task structures and constraints implicit in multi-step demonstrations. Specifically, we borrow from manipulation planning literature the concept of mode families, which group robot configurations by specific motion constraints, to serve as an abstraction layer between the high-level language representations of an LLM and the low-level physical trajectories of a robot. By replaying a few human demonstrations with synthetic perturbations, we generate coverage over the demonstrations' state space with additional successful executions as well as counterfactuals that fail the task. Our explanation-based learning framework trains an end-to-end differentiable neural network to predict successful trajectories from failures and as a by-product learns classifiers that ground low-level states and images in mode families without dense labeling. The learned grounding classifiers can further be used to translate language plans into reactive policies in the physical domain in an interpretable manner. We show our approach improves the interpretability and reactivity of imitation learning through 2D navigation and simulated and real robot manipulation tasks. Website: https://sites.google.com/view/grounding-plans

☆ Vision-Based Dexterous Motion Planning by Dynamic Movement Primitives with Human Hand Demonstration

Nuo Chen, Ya-Jun Pan

This paper proposes a vision-based framework for a 7-degree-of-freedom robotic manipulator, with the primary objective of facilitating its capacity to acquire information from human hand demonstrations for the execution of dexterous pick-and-place tasks. Most existing works only focus on the position demonstration without considering the orientations. In this paper, by employing a single depth camera, MediaPipe is applied to generate the three-dimensional coordinates of a human hand, thereby comprehensively recording the hand's motion, encompassing the trajectory of the wrist, orientation of the hand, and the grasp motion. A mean filter is applied during data pre-processing to smooth the raw data. The demonstration is designed to pick up an object at a specific angle, navigate around obstacles in its path and subsequently, deposit it within a sloped container. The robotic system demonstrates its learning capabilities, facilitated by the implementation of Dynamic Movement Primitives, enabling the assimilation of user actions into its trajectories with different start and end poi

☆ Berry Twist: a Twisting-Tube Soft Robotic Gripper for Blackberry Harvesting

Johannes F. Elfferich, Ebrahim Shahabi, Cosimo Della Santina, Dimitra Dodou

As global demand for fruits and vegetables continues to rise, the agricultural industry faces challenges in securing adequate labor. Robotic harvesting devices offer a promising solution to solve this issue. However, harvesting delicate fruits, notably blackberries, poses unique challenges due to their fragility. This study introduces and evaluates a prototype robotic gripper specifically designed for blackberry harvesting. The gripper features an innovative fabric tube mechanism employing motorized twisting action to gently envelop the fruit, ensuring uniform pressure application and minimizing damage. Three types of tubes were developed, varying in elasticity and compressibility using foam padding, spandex, and food-safe cotton cheesecloth. Performance testing focused on assessing each gripper's ability to detach and release blackberries, with emphasis on quantifying damage rates. Results indicate the proposed gripper achieved an 82% success rate in detaching blackberries and a 95% success rate in releasing them, showcasing the promised potential for robotic harvesting applications.

☆ A Comparative Analysis of Visual Odometry in Virtual and Real-World Railways Environments

Gianluca D'Amico, Mauro Marinoni, Giorgio Buttazzo

Perception tasks play a crucial role in the development of automated operations and systems across multiple application fields. In the railway transportation domain, these tasks can improve the safety, reliability, and efficiency of various perations, including train localization, signal recognition, and track discrimination. However, collecting considerable and precisely labeled datasets for testing such novel algorithms poses extreme challenges in the railway environment due to the severe restrictions in accessing the infrastructures and the practical difficulties associated with properly equipping trains with the required sensors, such as cameras and LiDARs. The remarkable innovations of graphic engine tools offer new solutions to craft realistic synthetic datasets. To illustrate the advantages of employing graphic simulation for early-stage testing of perception tasks in the railway domain, this paper presents a comparative analysis of the performance of a SLAM algorithm applied both in a virtual synthetic environment and a real-world scenario. The analysis leverages virtual railway environments created with the latest version of Unreal Engine, facilitating data collection and allowing the examination of challenging scenarios, including low-visibility, dangerous operational modes, and complex environments. The results highlight the feasibility and potentiality of graphic simulation to advance perception tasks in the railway domain.

☆ Trajectory Optimization with Global Yaw Parameterization for Field-of-View Constrained Autonomous Flight

Yuwei Wu, Yuezhan Tao, Igor Spasojevic, Vijay Kumar

Trajectory generation for quadrotors with limited field-of-view sensors has numerous applications such as aerial exploration, coverage, inspection, videography, and target tracking. Most previous works simplify the task of optimizing yaw trajectories by either aligning the heading of the robot with its velocity, or potentially restricting the feasible space of candidate trajectories by using a limited yaw domain to circumvent angular singularities. In this paper, we propose a novel \textit{global} yaw parameterization method for trajectory optimization that allows a 360-degree yaw variation as demanded by the underlying algorithm. This approach effectively bypasses inherent singularities by including supplementary quadratic constraints and transforming the final decision variables into the desired state representation. This method significantly reduces the needed control effort, and improves optimization feasibility. Furthermore, we apply the method to several examples of different applications that require jointly optimizing over both the yaw and position trajectories. Ultimately, we present a comprehensive numerical analysis and evaluation of our proposed method in both simulation and real-world experiments.

☆ Calib3D: Calibrating Model Preferences for Reliable 3D Scene Understanding

Lingdong Kong, Xiang Xu, Jun Cen, Wenwei Zhang, Liang Pan, Kai Chen, Ziwei Liu

Safety-critical 3D scene understanding tasks necessitate not only accurate but also confident predictions from 3D perception models. This study introduces Calib3D, a pioneering effort to benchmark and scrutinize the reliability of 3D scene understanding models from an uncertainty estimation viewpoint. We comprehensively evaluate 28 state-of-the-art models across 10 diverse 3D datasets, uncovering insightful phenomena that cope with both the aleatoric and epistemic uncertainties in 3D scene understanding. We discover that despite achieving impressive levels of accuracy, existing models frequently fail to provide reliable uncertainty estimates -- a pitfall that critically undermines their applicability in safety-sensitive contexts. Through extensive analysis of key factors such as network capacity, LiDAR representations, rasterization resolutions, and 3D data augmentation techniques, we correlate these aspects directly with the model calibration efficacy. Furthermore, we introduce DeptS, a novel depth-aware scaling approach aimed at enhancing 3D model calibration. Extensive experiments across a wide range of configurations validate the superiority of our method. We hope this work could serve as a cornerstone for fostering reliable 3D scene understanding. Code and benchmark toolkits are publicly available.

comment: Preprint; 37 pages, 8 figures, 11 tables; Code at https://github.com/ldkong1205/Calib3D

☆ Optimizing LiDAR Placements for Robust Driving Perception in Adverse Conditions

Ye Li, Lingdong Kong, Hanjiang Hu, Xiaohao Xu, Xiaonan Huang

The robustness of driving perception systems under unprecedented conditions is crucial for safety-critical usages. Latest advancements have prompted increasing interests towards multi-LiDAR perception. However, prevailing driving datasets predominantly utilize single-LiDAR systems and collect data devoid of adverse conditions, failing to capture the complexities of real-world environments accurately. Addressing these gaps, we proposed Place3D, a full-cycle pipeline that encompasses LiDAR placement optimization, data generation, and downstream evaluations. Our framework makes three appealing contributions. 1) To identify the most effective configurations for multi-LiDAR systems, we introduce a Surrogate Metric of the Semantic Occupancy Grids (M-SOG) to evaluate LiDAR placement quality. 2) Leveraging the M-SOG metric, we propose a novel optimization strategy to refine multi-LiDAR placements. 3) Centered around the theme of multi-condition multi-LiDAR perception, we collect a 364,000-frame dataset from both clean and adverse conditions. Extensive experiments demonstrate that LiDAR placements optimized using our approach outperform various baselines. We showcase exceptional robustness in both 3D object detection and LiDAR semantic segmentation tasks, under diverse adverse weather and sensor failure conditions. Code and benchmark toolkit are publicly available.

comment: Preprint; 40 pages, 11 figures, 15 tables; Code at https://github.com/ywyeli/Place3D

☆ DriveCoT: Integrating Chain-of-Thought Reasoning with End-to-End Driving

Tianqi Wang, Enze Xie, Ruihang Chu, Zhenguo Li, Ping Luo

End-to-end driving has made significant progress in recent years, demonstrating benefits such as system simplicity and competitive driving performance under both open-loop and closed-loop settings. Nevertheless, the lack of interpretability and controllability in its driving decisions hinders real-world deployment for end-to-end driving systems. In this paper, we collect a comprehensive end-to-end driving dataset named DriveCoT, leveraging the CARLA simulator. It contains sensor data, control decisions, and chain-of-thought labels to indicate the reasoning process. We utilize the challenging driving scenarios from the CARLA leaderboard 2.0, which involve high-speed driving and lane-changing, and propose a rule-based expert policy to control the vehicle and generate ground truth labels for its reasoning process across different driving aspects and the final decisions. This dataset can serve as an open-loop end-to-end driving benchmark, enabling the evaluation of accuracy in various chain-of-thought aspects and the final decision. In addition, we propose a baseline model called DriveCoT-Agent, trained on our dataset, to generate chain-of-thought predictions and final decisions. The trained model exhibits strong performance in both open-loop and closed-loop evaluations, demonstrating the effectiveness of our proposed dataset.

☆ Bayesian Methods for Trust in Collaborative Multi-Agent Autonomy

R. Spencer Hallyburton, Miroslav Pajic

Multi-agent, collaborative sensor fusion is a vital component of a multi-national intelligence toolkit. In safety-critical and/or contested environments, adversaries may infiltrate and compromise a number of agents. We analyze state of the art multi-target tracking algorithms under this compromised agent threat model. We prove that the track existence probability test ("track score") is significantly vulnerable to even small numbers of adversaries. To add security awareness, we design a trust estimation framework using hierarchical Bayesian updating. Our framework builds beliefs of trust on tracks and agents by mapping sensor measurements to trust pseudomeasurements (PSMs) and incorporating prior trust beliefs in a Bayesian context. In case studies, our trust estimation algorithm accurately estimates the trustworthiness of tracks/agents, subject to observability limitations.

☆ Learning Symbolic and Subsymbolic Temporal Task Constraints from Bimanual Human Demonstrations IROS 2024

Christian Dreher, Tamim Asfour

Learning task models of bimanual manipulation from human demonstration and their execution on a robot should take temporal constraints between actions into account. This includes constraints on (i) the symbolic level such as precedence relations or temporal overlap in the execution, and (ii) the subsymbolic level such as the duration of different actions, or their starting and end points in time. Such temporal constraints are crucial for temporal planning, reasoning, and the exact timing for the execution of bimanual actions on a bimanual robot. In our previous work, we addressed the learning of temporal task constraints on the symbolic level and demonstrated how a robot can leverage this knowledge to respond to failures during execution. In this work, we propose a novel model-driven approach for the combined learning of symbolic and subsymbolic temporal task constraints from multiple bimanual human demonstrations. Our main contributions are a subsymbolic foundation of a temporal task model that describes temporal nexuses of actions in the task based on distributions of temporal differences between semantic action keypoints, as well as a method based on fuzzy logic to derive symbolic temporal task constraints from this representation. This complements our previous work on learning comprehensive temporal task models by integrating symbolic and subsymbolic information based on a subsymbolic foundation, while still maintaining the symbolic expressiveness of our previous approach. We compare our proposed approach with our previous pure-symbolic approach and show that we can reproduce and even outperform it. Additionally, we show how the subsymbolic temporal task constraints can synchronize otherwise unimanual movement primitives for bimanual behavior on a humanoid robot.

comment: 8 pages, submitted to IROS 2024

★ DHP-Mapping: A Dense Panoptic Mapping System with Hierarchical World Representation and Label Optimization Techniques IROS 2024

Tianshuai Hu, Jianhao Jiao, Yucheng Xu, Hongji Liu, Sheng Wang, Ming Liu

Maps provide robots with crucial environmental knowledge, thereby enabling them to perform interactive tasks effectively. Easily accessing accurate abstract-to-detailed geometric and semantic concepts from maps is crucial for robots to make informed and efficient decisions. To comprehensively model the environment and effectively manage the map data structure, we propose DHP-Mapping, a dense mapping system that utilizes multiple Truncated Signed Distance Field (TSDF) submaps and panoptic labels to hierarchically model the environment. The output map is able to maintain both voxel- and submap-level metric and semantic information. Two modules are presented to enhance the mapping efficiency and label consistency: (1) an inter-submaps label fusion strategy to eliminate duplicate points across submaps and (2) a conditional random field (CRF) based approach to enhance panoptic labels through object label comprehension and contextual information. We conducted experiments with two public datasets including indoor and outdoor scenarios. Our system performs comparably to state-of-the-art (SOTA) methods across geometry and label accuracy evaluation metrics. The experiment results highlight the effectiveness and scalability of our system, as it is capable of constructing precise geometry and maintaining consistent panoptic labels. Our code is publicly available at https://github.com/hutslib/DHP-Mapping.

comment: Submit to IROS 2024. Project website https://github.com/hutslib/DHP-Mapping

☆ Proprioception Is All You Need: Terrain Classification for Boreal Forests IROS 2024

Damien LaRocque, William Guimont-Martin, David-Alexandre Duclos, Philippe Giguère, François Pomerleau

Recent works in field robotics highlighted the importance of resiliency against different types of terrains. Boreal forests, in particular, are home to many mobility-impeding terrains that should be considered for off-road autonomous navigation. Also, being one of the largest land biomes on Earth, boreal forests are an area where autonomous vehicles are expected to become increasingly common. In this paper, we address this issue by introducing BorealTC, a publicly available dataset for proprioceptive-based terrain classification (TC). Recorded with a Husky A200, our dataset contains 116 min of Inertial Measurement Unit (IMU), motor current, and wheel odometry data, focusing on typical boreal forest terrains, notably snow, ice, and silty loam. Combining our dataset with another dataset from the state-of-the-art, we evaluate both a Convolutional Neural Network (CNN) and the novel state space model (SSM)-based Mamba architecture on a TC task. Interestingly, we show that while CNN outperforms Mamba on each separate dataset, Mamba achieves greater accuracy when trained on a combination of both. In addition, we demonstrate that Mamba's learning capacity is greater than a CNN for increasing amounts of data. We show that the combination of two TC datasets yields a latent space that can be interpreted with the properties of the terrains. We also discuss the implications of merging datasets on classification. Our source code and dataset are publicly available online: https://github.com/norlab-ulaval/BorealTC.

comment: Submitted to the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024)

☆ TAIL: A Terrain-Aware Multi-Modal SLAM Dataset for Robot Locomotion in Deformable Granular Environments

Chen Yao, Yangtao Ge, Guowei Shi, Zirui Wang, Ningbo Yang, Zheng Zhu, Hexiang Wei, Yuntian Zhao, Jing Wu, Zhenzhong Jia

Terrain-aware perception holds the potential to improve the robustness and accuracy of autonomous robot navigation in the wilds, thereby facilitating effective off-road traversals. However, the lack of multi-modal perception across various motion patterns hinders the solutions of Simultaneous Localization And Mapping (SLAM), especially when confronting non-geometric hazards in demanding landscapes. In this paper, we first propose a Terrain-Aware multI-modaL (TAIL) dataset tailored to deformable and sandy terrains. It incorporates various types of robotic proprioception and distinct ground interactions for the unique challenges and benchmark of multi-sensor fusion SLAM. The versatile sensor suite comprises stereo frame cameras, multiple ground-pointing RGB-D cameras, a rotating 3D LiDAR, an IMU, and an RTK device. This ensemble is hardware-synchronized, well-calibrated, and self-contained. Utilizing both wheeled and quadrupedal locomotion, we efficiently collect comprehensive sequences to capture rich unstructured scenarios. It spans the spectrum of scope, terrain interactions, scene changes, ground-level properties, and dynamic robot characteristics. We benchmark several state-of-the-art SLAM methods against ground truth and provide performance validations. Corresponding challenges and limitations are also reported. All associated resources are accessible upon request at \url{https://tailrobot.github.io/}.

comment: Submitted to IEEE Robotics and Automation Letters

☆ A Semi-Lagrangian Approach for Time and Energy Path Planning Optimization in Static Flow Fields

Víctor C. da S. Campos, Armando A. Neto, Douglas G. Macharet

Efficient path planning for autonomous mobile robots is a critical problem across numerous domains, where optimizing both time and energy consumption is paramount. This paper introduces a novel methodology that considers the dynamic influence of an environmental flow field and considers geometric constraints, including obstacles and forbidden zones, enriching the complexity of the planning problem. We formulate it as a multi-objective optimal control problem, propose a novel transformation called Harmonic Transformation, and apply a semi-Lagrangian scheme to solve it. The set of Pareto efficient solutions is obtained considering two distinct approaches: a deterministic method and an evolutionary-based one, both of which are designed to make use of the proposed Harmonic Transformation. Through an extensive analysis of these approaches, we demonstrate their efficacy in finding optimized paths.

comment: 12 pages, initial paper submission; Preprint submitted to the IEEE Transactions on Intelligent Transportation Systems

☆ Exploiting Priors from 3D Diffusion Models for RGB-Based One-Shot View Planning IROS 2024

Sicong Pan, Liren Jin, Xuying Huang, Cyrill Stachniss, Marija Popović, Maren Bennewitz

Object reconstruction is relevant for many autonomous robotic tasks that require interaction with the environment. A key challenge in such scenarios is planning view configurations to collect informative measurements for reconstructing an initially unknown object. One-shot view planning enables efficient data collection by predicting view configurations and planning the globally shortest path connecting all views at once. However, geometric priors about the object are required to conduct one-shot view planning. In this work, we propose a novel one-shot view planning approach that utilizes the powerful 3D generation capabilities of diffusion models as priors. By incorporating such geometric priors into our pipeline, we achieve effective one-shot view planning starting with only a single RGB image of the object to be reconstructed. Our planning experiments in simulation and real-world setups indicate that our approach balances well between object reconstruction quality and movement cost.

comment: Sicong Pan and Liren Jin have equal contribution. Submitted to IROS 2024

★ CurbNet: Curb Detection Framework Based on LiDAR Point Cloud Segmentation

Guoyang Zhao, Fulong Ma, Yuxuan Liu, Weiqing Qi, Ming Liu

Curb detection is an important function in intelligent driving and can be used to determine drivable areas of the road. However, curbs are difficult to detect due to the complex road environment. This paper introduces CurbNet, a novel framework for curb detection, leveraging point cloud segmentation. Addressing the dearth of comprehensive curb datasets and the absence of 3D annotations, we have developed the 3D-Curb dataset, encompassing 7,100 frames, which represents the largest and most categorically diverse collection of curb point clouds currently available. Recognizing that curbs are primarily characterized by height variations, our approach harnesses spatially-rich 3D point clouds for training. To tackle the challenges presented by the uneven distribution of curb features on the xy-plane and their reliance on z-axis high-frequency features, we introduce the multi-scale and channel attention (MSCA) module, a bespoke solution designed to optimize detection performance. Moreover, we propose an adaptive weighted loss function group, specifically formulated to counteract the imbalance in the distribution of curb point clouds relative to other categories. Our extensive experimentation on 2 major datasets has yielded results that surpass existing benchmarks set by leading curb detection and point cloud segmentation models. By integrating multi-clustering and curve fitting techniques in our post-processing stage, we have substantially reduced noise in curb detection, thereby enhancing precision to 0.8744. Notably, CurbNet has achieved an exceptional average metrics of over 0.95 at a tolerance of just 0.15m, thereby establishing a new benchmark. Furthermore, corroborative real-world experiments and dataset analyzes mutually validate each other, solidifying CurbNet's superior detection proficiency and its robust generalizability.

☆ DBPF: A Framework for Efficient and Robust Dynamic Bin-Picking

Yichuan Li, Junkai Zhao, Yixiao Li, Zheng Wu, Rui Cao, Masayoshi Tomizuka, Yunhui Liu

Efficiency and reliability are critical in robotic bin-picking as they directly impact the productivity of automated industrial processes. However, traditional approaches, demanding static objects and fixed collisions, lead to deployment limitations, operational inefficiencies, and process unreliability. This paper introduces a Dynamic Bin-Picking Framework (DBPF) that challenges traditional static assumptions. The DBPF endows the robot with the reactivity to pick multiple moving arbitrary objects while avoiding dynamic obstacles, such as the moving bin. Combined with scene-level pose generation, the proposed pose selection metric leverages the Tendency-Aware Manipulability Network optimizing suction pose determination. Heuristic task-specific designs like velocity-matching, dynamic obstacle avoidance, and the resight policy, enhance the picking success rate and reliability. Empirical experiments demonstrate the importance of these components. Our method achieves an average 84% success rate, surpassing the 60% of the most comparable baseline, crucially, with zero collisions. Further evaluations under diverse dynamic scenarios showcase DBPF's robust performance in dynamic bin-picking. Results suggest that our framework offers a promising solution for efficient and reliable robotic bin-picking under dynamics.

comment: 8 pages, 5 figures. This paper has been accepted by IEEE RA-L on 2024-03-24. See the supplementary video at youtube: https://youtu.be/n5af2VsKhkg

☆ Visual Action Planning with Multiple Heterogeneous Agents

Martina Lippi, Michael C. Welle, Marco Moletta, Alessandro Marino, Andrea Gasparri, Danica Kragic

Visual planning methods are promising to handle complex settings where extracting the system state is challenging. However, none of the existing works tackles the case of multiple heterogeneous agents which are characterized by different capabilities and/or embodiment. In this work, we propose a method to realize visual action planning in multi-agent settings by exploiting a roadmap built in a low-dimensional structured latent space and used for planning. To enable multi-agent settings, we infer possible parallel actions from a dataset composed of tuples associated with individual actions. Next, we evaluate feasibility and cost of them based on the capabilities of the multi-agent system and endow the roadmap with this information, building a capability latent space roadmap (C-LSR). Additionally, a capability suggestion strategy is designed to inform the human operator about possible missing capabilities when no paths are found. The approach is validated in a simulated burger cooking task and a real-world box packing task.

☆ Low-Cost Teleoperation with Haptic Feedback through Vision-based Tactile Sensors for Rigid and Soft Object Manipulation

Martina Lippi, Michael C. Welle, Maciej K. Wozniak, Andrea Gasparri, Danica Kragic

Haptic feedback is essential for humans to successfully perform complex and delicate manipulation tasks. A recent rise in tactile sensors has enabled robots to leverage the sense of touch and expand their capability drastically. However, many tasks still need human intervention/guidance. For this reason, we present a teleoperation framework designed to provide haptic feedback to human operators based on the data from camera-based tactile sensors mounted on the robot gripper. Partial autonomy is introduced to prevent slippage of grasped objects during task execution. Notably, we rely exclusively on low-cost off-the-shelf hardware to realize an affordable solution. We demonstrate the versatility of the framework on nine different objects ranging from rigid to soft and fragile ones, using three different operators on real hardware.

comment: https://vision-tactile-manip.github.io/teleop/

☆ ProIn: Learning to Predict Trajectory Based on Progressive Interactions for Autonomous Driving

Yinke Dong, Haifeng Yuan, Hongkun Liu, Wei Jing, Fangzhen Li, Hongmin Liu, Bin Fan

Accurate motion prediction of pedestrians, cyclists, and other surrounding vehicles (all called agents) is very important for autonomous driving. Most existing works capture map information through an one-stage interaction with map by vector-based attention, to provide map constraints for social interaction and multi-modal differentiation. However, these methods have to encode all required map rules into the focal agent's feature, so as to retain all possible intentions' paths while at the meantime to adapt to potential social interaction. In this work, a progressive interaction network is proposed to enable the agent's feature to progressively focus on relevant maps, in order to better learn agents' feature representation capturing the relevant map constraints. The network progressively encode the complex influence of map constraints into the agent's feature through graph convolutions at the following three stages: after historical trajectory encoder, after social interaction, and after multi-modal differentiation. In addition, a weight allocation mechanism is proposed for multi-modal training, so that each mode can obtain learning opportunities from a single-mode ground truth. Experiments have validated the superiority of progressive interactions to the existing one-stage interaction, and demonstrate the effectiveness of each component. Encouraging results were obtained in the challenging benchmarks.

♻ ☆ A Modular Pneumatic Soft Gripper Design for Aerial Grasping and Landing

Hiu Ching Cheung, Ching-Wei Chang, Bailun Jiang, Chih-Yung Wen, Henry K. Chu

Aerial robots have garnered significant attention due to their potential applications in various industries, such as inspection, search and rescue, and drone delivery. Successful missions often depend on the ability of these robots to grasp and land effectively. This paper presents a novel modular soft gripper design tailored explicitly for aerial grasping and landing operations. The proposed modular pneumatic soft gripper incorporates a feed-forward proportional controller to regulate pressure, enabling compliant gripping capabilities. The modular connectors of the soft fingers offer two configurations for the 4-tip soft gripper, H-base (cylindrical) and X-base (spherical), allowing adaptability to different target objects. Additionally, the gripper can serve as a soft landing gear when deflated, eliminating the need for an extra landing gear. This design reduces weight, simplifies aerial manipulation control, and enhances flight efficiency. We demonstrate the efficacy of indoor aerial grasping and achieve a maximum payload of 217 g using the proposed soft aerial vehicle and its H-base pneumatic soft gripper (808 g).

comment: 7 pages, 13 figures, accepted by IEEE RoboSoft 2024

♻ ☆ Robust Integral Consensus Control of Multi-Agent Networks Perturbed by Matched and Unmatched Disturbances: The Case of Directed Graphs

Jose Guadalupe Romero, David Navarro-Alarcon

This work presents a new method to design consensus controllers for perturbed double integrator systems whose interconnection is described by a directed graph containing a rooted spanning tree. We propose new robust controllers to solve the consensus and synchronization problems when the systems are under the effects of matched and unmatched disturbances. In both problems, we present simple continuous controllers, whose integral actions allow us to handle the disturbances. A rigorous stability analysis based on Lyapunov's direct method for unperturbed networked systems is presented. To assess the performance of our result, a representative simulation study is presented.

♻ ★ A Group Theoretic Metric for Robot State Estimation Leveraging Chebyshev Interpolation ICRA 2024

Varun Agrawal, Frank Dellaert

We propose a new metric for robot state estimation based on the recently introduced $\text{SE}_2(3)$ Lie group definition. Our metric is related to prior metrics for SLAM but explicitly takes into account the linear velocity of the state estimate, improving over current pose-based trajectory analysis. This has the benefit of providing a single, quantitative metric to evaluate state estimation algorithms against, while being compatible with existing tools and libraries. Since ground truth data generally consists of pose data from motion capture systems, we also propose an approach to compute the ground truth linear velocity based on polynomial interpolation. Using Chebyshev interpolation and a pseudospectral parameterization, we can accurately estimate the ground truth linear velocity of the trajectory in an optimal fashion with best approximation error. We demonstrate how this approach performs on multiple robotic platforms where accurate state estimation is vital, and compare it to alternative approaches such as finite differences. The pseudospectral parameterization also provides a means of trajectory data compression as an additional benefit. Experimental results show our method provides a valid and accurate means of comparing state estimation systems, which is also easy to interpret and report.

comment: Accepted to ICRA 2024

♻ ☆ I-PHYRE: Interactive Physical Reasoning ICLR 2024

Shiqian Li, Kewen Wu, Chi Zhang, Yixin Zhu

Current evaluation protocols predominantly assess physical reasoning in stationary scenes, creating a gap in evaluating agents' abilities to interact with dynamic events. While contemporary methods allow agents to modify initial scene configurations and observe consequences, they lack the capability to interact with events in real time. To address this, we introduce I-PHYRE, a framework that challenges agents to simultaneously exhibit intuitive physical reasoning, multi-step planning, and in-situ intervention. Here, intuitive physical reasoning refers to a quick, approximate understanding of physics to address complex problems; multi-step denotes the need for extensive sequence planning in I-PHYRE, considering each intervention can significantly alter subsequent choices; and in-situ implies the necessity for timely object manipulation within a scene, where minor timing deviations can result in task failure. We formulate four game splits to scrutinize agents' learning and generalization of essential principles of interactive physical reasoning, fostering learning through interaction with representative scenarios. Our exploration involves three planning strategies and examines several supervised and reinforcement agents' zero-shot generalization proficiency on I-PHYRE. The outcomes highlight a notable gap between existing learning algorithms and human performance, emphasizing the imperative for more research in enhancing agents with interactive physical reasoning capabilities. The environment and baselines will be made publicly available.

comment: 21 pages, ICLR 2024

♻ ☆ OpenFMNav: Towards Open-Set Zero-Shot Object Navigation via Vision-Language Foundation Models NAACL 2024

Yuxuan Kuang, Hai Lin, Meng Jiang

Object navigation (ObjectNav) requires an agent to navigate through unseen environments to find queried objects. Many previous methods attempted to solve this task by relying on supervised or reinforcement learning, where they are trained on limited household datasets with close-set objects. However, two key challenges are unsolved: understanding free-form natural language instructions that demand open-set objects, and generalizing to new environments in a zero-shot manner. Aiming to solve the two challenges, in this paper, we propose OpenFMNav, an Open-set Foundation Model based framework for zero-shot object Navigation. We first unleash the reasoning abilities of large language models (LLMs) to extract proposed objects from natural language instructions that meet the user's demand. We then leverage the generalizability of large vision language models (VLMs) to actively discover and detect candidate objects from the scene, building a Versatile Semantic Score Map (VSSM). Then, by conducting common sense reasoning on VSSM, our method can perform effective language-guided exploration and exploitation of the scene and finally reach the goal. By leveraging the reasoning and generalizing abilities of foundation models, our method can understand free-form human instructions and perform effective open-set zero-shot navigation in diverse environments. Extensive experiments on the HM3D ObjectNav benchmark show that our method surpasses all the strong baselines on all metrics, proving our method's effectiveness. Furthermore, we perform real robot demonstrations to validate our method's open-set-ness and generalizability to real-world environments.

comment: NAACL 2024 Findings

♻ ☆ Directionality-Aware Mixture Model Parallel Sampling for Efficient Linear Parameter Varying Dynamical System Learning

Sunan Sun, Haihui Gao, Tianyu Li, Nadia Figueroa

The Linear Parameter Varying Dynamical System (LPV-DS) is an effective approach that learns stable, time-invariant motion policies using statistical modeling and semi-definite optimization to encode complex motions for reactive robot control. Despite its strengths, the LPV-DS learning approach faces challenges in achieving a high model accuracy without compromising the computational efficiency. To address this, we introduce the Directionality-Aware Mixture Model (DAMM), a novel statistical model that applies the Riemannian metric on the n-sphere $\mathbb{S}^n$ to efficiently blend non-Euclidean directional data with $\mathbb{R}^m$ Euclidean states. Additionally, we develop a hybrid Markov chain Monte Carlo technique that combines Gibbs Sampling with Split/Merge Proposal, allowing for parallel computation to drastically speed up inference. Our extensive empirical tests demonstrate that LPV-DS integrated with DAMM achieves higher reproduction accuracy, better model efficiency, and near real-time/online learning compared to standard estimation methods on various datasets. Lastly, we demonstrate its suitability for incrementally learning multi-behavior policies in real-world robot experiments.

♻ ☆ DTG : Diffusion-based Trajectory Generation for Mapless Global Navigation

Jing Liang, Amirreza Payandeh, Daeun Song, Xuesu Xiao, Dinesh Manocha

We present a novel end-to-end diffusion-based trajectory generation method, DTG, for mapless global navigation in challenging outdoor scenarios with occlusions and unstructured off-road features like grass, buildings, bushes, etc. Given a distant goal, our approach computes a trajectory that satisfies the following goals: (1) minimize the travel distance to the goal; (2) maximize the traversability by choosing paths that do not lie in undesirable areas. Specifically, we present a novel Conditional RNN(CRNN) for diffusion models to efficiently generate trajectories. Furthermore, we propose an adaptive training method that ensures that the diffusion model generates more traversable trajectories. We evaluate our methods in various outdoor scenes and compare the performance with other global navigation algorithms on a Husky robot. In practice, we observe at least a 15% improvement in traveling distance and around a 7% improvement in traversability.

comment: 10 pages

♻ ☆ GelLink: A Compact Multi-phalanx Finger with Vision-based Tactile Sensing and Proprioception ICRA 2024

Yuxiang Ma, Jialiang Zhao, Edward Adelson

Compared to fully-actuated robotic end-effectors, underactuated ones are generally more adaptive, robust, and cost-effective. However, state estimation for underactuated hands is usually more challenging. Vision-based tactile sensors, like Gelsight, can mitigate this issue by providing high-resolution tactile sensing and accurate proprioceptive sensing. As such, we present GelLink, a compact, underactuated, linkage-driven robotic finger with low-cost, high-resolution vision-based tactile sensing and proprioceptive sensing capabilities. In order to reduce the amount of embedded hardware, i.e. the cameras and motors, we optimize the linkage transmission with a planar linkage mechanism simulator and develop a planar reflection simulator to simplify the tactile sensing hardware. As a result, GelLink only requires one motor to actuate the three phalanges, and one camera to capture tactile signals along the entire finger. Overall, GelLink is a compact robotic finger that shows adaptability and robustness when performing grasping tasks. The integration of vision-based tactile sensors can significantly enhance the capabilities of underactuated fingers and potentially broaden their future usage.

comment: Supplement video: https://www.youtube.com/watch?v=hZwUpAig5C0 . 7 pages, 9 figures. ICRA 2024 (IEEE International Conference on Robotics and Automation)

♻ ☆ Fast LiDAR Informed Visual Search in Unseen Indoor Environments

Ryan Gupta, Kyle Morgenstein, Steven Ortega, Luis Sentis

This paper explores the problem of planning for visual search without prior map information. We leverage the pixel-wise environment perception problem where one is given wide Field of View 2D scan data and must perform LiDAR segmentation to contextually label points in the surroundings. These pixel classifications provide an informed prior on which to plan next best viewpoints during visual search tasks. We present LIVES: LiDAR Informed Visual Search, a method aimed at finding objects of interest in unknown indoor environments. A robust map-free classifier is trained from expert data collected using a simple cart platform equipped with a map-based classifier. An autonomous exploration planner takes the contextual data from scans and uses that prior to plan viewpoints more likely to yield detection of the search target. We propose a utility function that accounts for traditional metrics like information gain and path cost and for the contextual information. LIVES is baselined against several existing exploration methods in simulation to verify its performance. It is validated in real-world experiments with single and multiple search objects with a Spot robot in two unseen environments. Videos of experiments, implementation details and open source code can be found at https://sites.google.com/view/lives-2024/home.

comment: 6 pages + references. 6 figures. 1 algorithm. 1 table

♻ ☆ Diagrammatic Instructions to Specify Spatial Objectives and Constraints with Applications to Mobile Base Placement

Qilin Sun, Weiming Zhi, Tianyi Zhang, Matthew Johnson-Roberson

This paper introduces Spatial Diagrammatic Instructions (SDIs), an approach for human operators to specify objectives and constraints that are related to spatial regions in the working environment. Human operators are enabled to sketch out regions directly on camera images that correspond to the objectives and constraints. These sketches are projected to 3D spatial coordinates, and continuous Spatial Instruction Maps (SIMs) are learned upon them. These maps can then be integrated into optimization problems for tasks of robots. In particular, we demonstrate how Spatial Diagrammatic Instructions can be applied to solve the Base Placement Problem of mobile manipulators, which concerns the best place to put the manipulator to facilitate a certain task. Human operators can specify, via sketch, spatial regions of interest for a manipulation task and permissible regions for the mobile manipulator to be at. Then, an optimization problem that maximizes the manipulator's reachability, or coverage, over the designated regions of interest while remaining in the permissible regions is solved. We provide extensive empirical evaluations, and show that our formulation of Spatial Instruction Maps provides accurate representations of user-specified diagrammatic instructions. Furthermore, we demonstrate that our diagrammatic approach to the Mobile Base Placement Problem enables higher quality solutions and faster run-time.

♻ ☆ Greedy Perspectives: Multi-Drone View Planning for Collaborative Perception in Cluttered Environments IROS'24

Krishna Suresh, Aditya Rauniyar, Micah Corah, Sebastian Scherer

Deployment of teams of aerial robots could enable large-scale filming of dynamic groups of people (actors) in complex environments for applications in areas such as team sports and cinematography. Toward this end, methods for submodular maximization via sequential greedy planning can be used for scalable optimization of camera views across teams of robots but face challenges with efficient coordination in cluttered environments. Obstacles can produce occlusions and increase chances of inter-robot collision which can violate requirements for near-optimality guarantees. To coordinate teams of aerial robots in filming groups of people in dense environments, a more general view-planning approach is required. We explore how collision and occlusion impact performance in filming applications through the development of a multi-robot multi-actor view planner with an occlusion-aware objective for filming groups of people and compare with a formation planner and a greedy planner that ignores inter-robot collisions. We evaluate our approach based on five test environments and complex multi-actor behaviors. Compared with a formation planner, our sequential planner generates 14% greater view reward over the actors for three scenarios and comparable performance to formation planning on two others. We also observe near identical view rewards for sequential planning both with and without inter-robot collision constraints which indicates that robots are able to avoid collisions without impairing performance in the perception task. Overall, we demonstrate effective coordination of teams of aerial robots for filming groups that may split, merge, or spread apart and in environments cluttered with obstacles that may cause collisions or occlusions.

comment: Submitted to IROS'24; 8 pages, 8 figures, 2 tables

♻ ☆ Optimizing Crowd-Aware Multi-Agent Path Finding through Local Broadcasting with Graph Neural Networks

Phu Pham, Aniket Bera

Multi-Agent Path Finding (MAPF) in crowded environments presents a challenging problem in motion planning, aiming to find collision-free paths for all agents in the system. MAPF finds a wide range of applications in various domains, including aerial swarms, autonomous warehouse robotics, and self-driving vehicles. Current approaches to MAPF generally fall into two main categories: centralized and decentralized planning. Centralized planning suffers from the curse of dimensionality when the number of agents or states increases and thus does not scale well in large and complex environments. On the other hand, decentralized planning enables agents to engage in real-time path planning within a partially observable environment, demonstrating implicit coordination. However, they suffer from slow convergence and performance degradation in dense environments. In this paper, we introduce CRAMP, a novel crowd-aware decentralized reinforcement learning approach to address this problem by enabling efficient local communication among agents via Graph Neural Networks (GNNs), facilitating situational awareness and decision-making capabilities in congested environments. We test CRAMP on simulated environments and demonstrate that our method outperforms the state-of-the-art decentralized methods for MAPF on various metrics. CRAMP improves the solution quality up to 59% measured in makespan and collision count, and up to 35% improvement in success rate in comparison to previous methods.

comment: 8 pages, 5 figures, 2 tables

♻ ☆ SO(2)-Equivariant Downwash Models for Close Proximity Flight

H. Smith, A. Shankar, J. Gielis, J. Blumenkamp, A. Prorok

Multirotors flying in close proximity induce aerodynamic wake effects on each other through propeller downwash. Conventional methods have fallen short of providing adequate 3D force-based models that can be incorporated into robust control paradigms for deploying dense formations. Thus, learning a model for these downwash patterns presents an attractive solution. In this paper, we present a novel learning-based approach for modelling the downwash forces that exploits the latent geometries (i.e. symmetries) present in the problem. We demonstrate that when trained with only 5 minutes of real-world flight data, our geometry-aware model outperforms state-of-the-art baseline models trained with more than 15 minutes of data. In dense real-world flights with two vehicles, deploying our model online improves 3D trajectory tracking by nearly 36% on average (and vertical tracking by 56%).

♻ ☆ Decision-Oriented Learning Using Differentiable Submodular Maximization for Multi-Robot Coordination

Guangyao Shi, Chak Lam Shek, Nare Karapetyan, Pratap Tokekar

We present a differentiable, decision-oriented learning framework for cost prediction in a class of multi-robot decision-making problems, in which the robots need to trade off the task performance with the costs of taking actions when they select actions to take. Specifically, we consider the cases where the task performance is measured by a known monotone submodular function (e.g., coverage, mutual information), and the cost of actions depends on the context (e.g., wind and terrain conditions). We need to learn a function that maps the context to the costs. Classically, we treat such a learning problem and the downstream decision-making problem as two decoupled problems, i.e., we first learn to predict the cost function without considering the downstream decision-making problem, and then use the learned function for predicting the cost and using it in the decision-making problem. However, the loss function used in learning a prediction function may not be aligned with the downstream decision-making. We propose a decision-oriented learning framework that incorporates the downstream task performance in the prediction phase via a differentiable optimization layer. The main computational challenge in such a framework is to make the combinatorial optimization, i.e., non-monotone submodular maximization, differentiable. This function is not naturally differentiable. We propose the Differentiable Cost Scaled Greedy algorithm (D-CSG), which is a continuous and differentiable relaxation of CSG. We demonstrate the efficacy of the proposed framework through numerical simulations. The results show that the proposed framework can result in better performance than the traditional two-stage approach.

comment: arXiv admin note: text overlap with arXiv:2303.01543

♻ ☆ Pre-Trained Masked Image Model for Mobile Robot Navigation ICRA 2024

Vishnu Dutt Sharma, Anukriti Singh, Pratap Tokekar

2D top-down maps are commonly used for the navigation and exploration of mobile robots through unknown areas. Typically, the robot builds the navigation maps incrementally from local observations using onboard sensors. Recent works have shown that predicting the structural patterns in the environment through learning-based approaches can greatly enhance task efficiency. While many such works build task-specific networks using limited datasets, we show that the existing foundational vision networks can accomplish the same without any fine-tuning. Specifically, we use Masked Autoencoders, pre-trained on street images, to present novel applications for field-of-view expansion, single-agent topological exploration, and multi-agent exploration for indoor mapping, across different input modalities. Our work motivates the use of foundational vision models for generalized structure prediction-driven applications, especially in the dearth of training data. For more qualitative results see https://raaslab.org/projects/MIM4Robots.

comment: Accepted at ICRA 2024

♻ ☆ C3D: Cascade Control with Change Point Detection and Deep Koopman Learning for Autonomous Surface Vehicles

Jianwen Li, Hyunsang Park, Wenjian Hao, Lei Xin, Jalil Chavez-Galaviz, Ajinkya Chaudhary, Meredith Bloss, Kyle Pattison, Christopher Vo, Devesh Upadhyay, Shreyas Sundaram, Shaoshuai Mou, Nina Mahmoudian

In this paper, we discuss the development and deployment of a robust autonomous system capable of performing various tasks in the maritime domain under unknown dynamic conditions. We investigate a data-driven approach based on modular design for ease of transfer of autonomy across different maritime surface vessel platforms. The data-driven approach alleviates issues related to a priori identification of system models that may become deficient under evolving system behaviors or shifting, unanticipated, environmental influences. Our proposed learning-based platform comprises a deep Koopman system model and a change point detector that provides guidance on domain shifts prompting relearning under severe exogenous and endogenous perturbations. Motion control of the autonomous system is achieved via an optimal controller design. The Koopman linearized model naturally lends itself to a linear-quadratic regulator (LQR) control design. We propose the C3D control architecture Cascade Control with Change Point Detection and Deep Koopman Learning. The framework is verified in station keeping task on an ASV in both simulation and real experiments. The approach achieved at least 13.9 percent improvement in mean distance error in all test cases compared to the methods that do not consider system changes.

comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

♻ ☆ On the Feedback Law in Stochastic Optimal Nonlinear Control

Mohamed Naveed Gul Mohamed, Suman Chakravorty, Raman Goyal, Ran Wang

We consider the problem of nonlinear stochastic optimal control. This problem is thought to be fundamentally intractable owing to Bellman's ``curse of dimensionality". We present a result that shows that repeatedly solving an open-loop deterministic problem from the current state with progressively shorter horizons, similar to Model Predictive Control (MPC), results in a feedback policy that is $O(\epsilon^4)$ near to the true global stochastic optimal policy, \nxx{where $\epsilon$ is a perturbation parameter modulating the noise.} We show that the optimal deterministic feedback problem has a perturbation structure in that higher-order terms of the feedback law do not affect lower-order terms, and that this structure is lost in the optimal stochastic feedback problem. Consequently, solving the Stochastic Dynamic Programming problem is highly susceptible to noise, even when tractable, and in practice, the MPC-type feedback law offers superior performance even for stochastic systems.

comment: arXiv admin note: substantial text overlap with arXiv:2002.10505, arXiv:2002.09478

♻ ☆ Body-mounted MR-conditional Robot for Minimally Invasive Liver Intervention

Zhefeng Huang, Anthony L. Gunderman, Samuel E. Wilcox, Saikat Sengupta, Jay Shah, Aiming Lu, David Woodrum, Yue Chen

MR-guided microwave ablation (MWA) has proven effective in treating hepatocellular carcinoma (HCC) with small-sized tumors, but the state-of-the-art technique suffers from sub-optimal workflow due to speed and accuracy of needle placement. This paper presents a compact body-mounted MR-conditional robot that can operate in closed-bore MR scanners for accurate needle guidance. The robotic platform consists of two stacked Cartesian XY stages, each with two degrees of freedom, that facilitate needle guidance. The robot is actuated using 3D-printed pneumatic turbines with MR-conditional bevel gear transmission systems. Pneumatic valves and control mechatronics are located inside the MRI control room and are connected to the robot with pneumatic transmission lines and optical fibers. Free space experiments indicated robot-assisted needle insertion error of 2.6$\pm$1.3 mm at an insertion depth of 80 mm. The MR-guided phantom studies were conducted to verify the MR-conditionality and targeting performance of the robot. Future work will focus on the system optimization and validations in animal trials.

comment: 10 figures

♻ ☆ Mind Meets Robots: A Review of EEG-Based Brain-Robot Interaction Systems

Yuchong Zhang, Nona Rajabi, Farzaneh Taleb, Andrii Matviienko, Yong Ma, Mårten Björkman, Danica Kragic

Brain-robot interaction (BRI) empowers individuals to control (semi-)automated machines through their brain activity, either passively or actively. In the past decade, BRI systems have achieved remarkable success, predominantly harnessing electroencephalogram (EEG) signals as the central component. This paper offers an up-to-date and exhaustive examination of 87 curated studies published during the last five years (2018-2023), focusing on identifying the research landscape of EEG-based BRI systems. This review aims to consolidate and underscore methodologies, interaction modes, application contexts, system evaluation, existing challenges, and potential avenues for future investigations in this domain. Based on our analysis, we present a BRI system model with three entities: Brain, Robot, and Interaction, depicting the internal relationships of a BRI system. We especially investigate the essence and principles on interaction modes between human brains and robots, a domain that has not yet been identified anywhere. We then discuss these entities with different dimensions encompassed. Within this model, we scrutinize and classify current research, reveal insights, specify challenges, and provide recommendations for future research trajectories in this field. Meanwhile, we envision our findings offer a design space for future human-robot interaction (HRI) research, informing the creation of efficient BRI frameworks.

♻ ☆ Reinforcement Learning with Options and State Representation

Ayoub Ghriss, Masashi Sugiyama, Alessandro Lazaric

The current thesis aims to explore the reinforcement learning field and build on existing methods to produce improved ones to tackle the problem of learning in high-dimensional and complex environments. It addresses such goals by decomposing learning tasks in a hierarchical fashion known as Hierarchical Reinforcement Learning. We start in the first chapter by getting familiar with the Markov Decision Process framework and presenting some of its recent techniques that the following chapters use. We then proceed to build our Hierarchical Policy learning as an answer to the limitations of a single primitive policy. The hierarchy is composed of a manager agent at the top and employee agents at the lower level. In the last chapter, which is the core of this thesis, we attempt to learn lower-level elements of the hierarchy independently of the manager level in what is known as the "Eigenoption". Based on the graph structure of the environment, Eigenoptions allow us to build agents that are aware of the geometric and dynamic properties of the environment. Their decision-making has a special property: it is invariant to symmetric transformations of the environment, allowing as a consequence to greatly reduce the complexity of the learning task.

comment: Master Thesis 2018, MVA ENS Paris-Saclay, Tokyo RIKEN AIP

♻ ☆ Zero-BEV: Zero-shot Projection of Any First-Person Modality to BEV Maps

Gianluca Monaci, Leonid Antsfeld, Boris Chidlovskii, Christian Wolf

Bird's-eye view (BEV) maps are an important geometrically structured representation widely used in robotics, in particular self-driving vehicles and terrestrial robots. Existing algorithms either require depth information for the geometric projection, which is not always reliably available, or are trained end-to-end in a fully supervised way to map visual first-person observations to BEV representation, and are therefore restricted to the output modality they have been trained for. In contrast, we propose a new model capable of performing zero-shot projections of any modality available in a first person view to the corresponding BEV map. This is achieved by disentangling the geometric inverse perspective projection from the modality transformation, eg. RGB to occupancy. The method is general and we showcase experiments projecting to BEV three different modalities: semantic segmentation, motion vectors and object bounding boxes detected in first person. We experimentally show that the model outperforms competing methods, in particular the widely used baseline resorting to monocular depth estimation.

♻ ☆ Accelerating Model Predictive Control for Legged Robots through Distributed Optimization

Lorenzo Amatucci, Giulio Turrisi, Angelo Bratta, Victor Barasuol, Claudio Semini

This paper presents a novel approach to enhance Model Predictive Control (MPC) for legged robots through Distributed Optimization. Our method focuses on decomposing the robot dynamics into smaller, parallelizable subsystems, and utilizing the Alternating Direction Method of Multipliers (ADMM) to ensure consensus among them. Each subsystem is managed by its own Optimal Control Problem, with ADMM facilitating consistency between their optimizations. This approach not only decreases the computational time but also allows for effective scaling with more complex robot configurations, facilitating the integration of additional subsystems such as articulated arms on a quadruped robot. We demonstrate, through numerical evaluations, the convergence of our approach on two systems with increasing complexity. In addition, we showcase that our approach converges towards the same solution when compared to a state-of-the-art centralized whole-body MPC implementation. Moreover, we quantitatively compare the computational efficiency of our method to the centralized approach, revealing up to a 75\% reduction in computational time. Overall, our approach offers a promising avenue for accelerating MPC solutions for legged robots, paving the way for more effective utilization of the computational performance of modern hardware.

♻ ☆ Towards Precise 3D Human Pose Estimation with Multi-Perspective Spatial-Temporal Relational Transformers IJCNN 2024

Jianbin Jiao, Xina Cheng, Weijie Chen, Xiaoting Yin, Hao Shi, Kailun Yang

3D human pose estimation captures the human joint points in three-dimensional space while keeping the depth information and physical structure. That is essential for applications that require precise pose information, such as human-computer interaction, scene understanding, and rehabilitation training. Due to the challenges in data collection, mainstream datasets of 3D human pose estimation are primarily composed of multi-view video data collected in laboratory environments, which contains rich spatial-temporal correlation information besides the image frame content. Given the remarkable self-attention mechanism of transformers, capable of capturing the spatial-temporal correlation from multi-view video datasets, we propose a multi-stage framework for 3D sequence-to-sequence (seq2seq) human pose detection. Firstly, the spatial module represents the human pose feature by intra-image content, while the frame-image relation module extracts temporal relationships and 3D spatial positional relationship features between the multi-perspective images. Secondly, the self-attention mechanism is adopted to eliminate the interference from non-human body parts and reduce computing resources. Our method is evaluated on Human3.6M, a popular 3D human pose detection dataset. Experimental results demonstrate that our approach achieves state-of-the-art performance on this dataset. The source code will be available at https://github.com/WUJINHUAN/3D-human-pose.

comment: Accepted to IJCNN 2024. The source code will be available at https://github.com/WUJINHUAN/3D-human-pose

♻ ☆ Uni-RLHF: Universal Platform and Benchmark Suite for Reinforcement Learning with Diverse Human Feedback ICLR 2024

Yifu Yuan, Jianye Hao, Yi Ma, Zibin Dong, Hebin Liang, Jinyi Liu, Zhixin Feng, Kai Zhao, Yan Zheng

Reinforcement Learning with Human Feedback (RLHF) has received significant attention for performing tasks without the need for costly manual reward design by aligning human preferences. It is crucial to consider diverse human feedback types and various learning methods in different environments. However, quantifying progress in RLHF with diverse feedback is challenging due to the lack of standardized annotation platforms and widely used unified benchmarks. To bridge this gap, we introduce Uni-RLHF, a comprehensive system implementation tailored for RLHF. It aims to provide a complete workflow from real human feedback, fostering progress in the development of practical problems. Uni-RLHF contains three packages: 1) a universal multi-feedback annotation platform, 2) large-scale crowdsourced feedback datasets, and 3) modular offline RLHF baseline implementations. Uni-RLHF develops a user-friendly annotation interface tailored to various feedback types, compatible with a wide range of mainstream RL environments. We then establish a systematic pipeline of crowdsourced annotations, resulting in large-scale annotated datasets comprising more than 15 million steps across 30+ popular tasks. Through extensive experiments, the results in the collected datasets demonstrate competitive performance compared to those from well-designed manual rewards. We evaluate various design choices and offer insights into their strengths and potential areas of improvement. We wish to build valuable open-source platforms, datasets, and baselines to facilitate the development of more robust and reliable RLHF solutions based on realistic human feedback. The website is available at https://uni-rlhf.github.io/.

comment: Published as a conference paper at ICLR 2024. The website is available at https://uni-rlhf.github.io/

♻ ☆ Vehicle Trajectory Tracking Through Magnetic Sensors: A Case Study of Two-lane Road

Xiaojiang Ren, Yan Wang, Yingfan Geng

Intelligent Transportation Systems (ITS) have a pressing need for efficient and reliable traffic surveillance solutions. This paper for the first time proposes a surveillance system that utilizes low-cost magnetic sensors for detecting and tracking vehicles continuously along the road. The system uses multiple sensors mounted along the roadside and lane boundaries to capture the movement of vehicles. Real-time measurement data is collected by base stations and processed to produce vehicle trajectories that include position, timestamp, and speed. To address the challenge of tracking vehicles continuously on a road network using a large amount of unlabeled magnetic sensor measurements, we first define a vehicle trajectory tracking problem. We then propose a graph-based data association algorithm to track each detected vehicle, and design a related online algorithm framework respectively. We finally validate the performance via both experimental simulation and real-world road deployment. The experimental results demonstrate that the proposed solution provides a cost-effective solution to capture the driving status of vehicles and on that basis form various traffic safety and efficiency applications.